Infection and Inflammation in Bovine: Staphylococcus Au Reus Mastitis

INFECTION AND INFLAMMATION IN BOVINE
STAPHYLOCOCCUS AUREUS MASTITIS
FIONA JANE YOUNG
For the Degree of MASTER OF VETERINARY SCIENCE
UNIVERSITY
of
GLASGOW
Department of Veterinary Clinical Studies

University of Glasgow Veterinary School
May, 1997
Fiona Jane Young, 1997

ProQuest Number: 11007663
All rights reserved
INFORMATION TO ALL USERS

The quality of this reproduction is d e p e n d e n t upon the quality of the copy subm itted.
In the unlikely e v e n t that the a u thor did not send a c o m p le te m anuscript

and there are missing pages, these will be noted. Also, if m aterial had to be rem oved,
a n o te will ind ica te the deletion.
uest
ProQuest 11007663
Published by ProQuest LLC(2018). C opyright of the Dissertation is held by the Author.
All rights reserved.

This work is protected against unauthorized copying under Title 17, United States C o d e
M icroform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346
I0T5S
[GLASGOW
pNivEfisirr
jjBKAPT
Abstract
The aim of this project was to study the dynamics of intramammary infection and
resulting inflammation in Staphylococcus aureus mastitis in cows. Four cows with
naturally occurring subclinical mastitis in one or more quarters, caused by S. aureus,
were used in this study.
The strains of S. aureus causing infection in the cows were determined by restriction
enzyme fragmentation pattern (REFP) analysis. Two cows were infected with one strain
of S. aureus, designated strain A, and two cows were infected with a different strain of S.
aureus, designated strain B. The cows were subsequently challenged by the
intramammary route in a novel cross-over design with large numbers (lOMO9) of either
the indigenous strain or the non-indigenous strain of S. aureus. Quarters which were
naturally infected with S. aureus and then challenged by the intramammary route with S.
aureus remained persistently infected three weeks following challenge in three of the four
cows, irrespective of the challenge strain. The fourth cow required treatment with
antibiotics due to an increase in the severity of mastitis. The results of this study showed
that in one cow, intramammary challenge of a subclinically infected quarter with the
indigenous strain of S. aureus failed to induce an anamnestic immune response which
might have instigated clearance of the original infection from the quarter; while in the
other cow, the indigenous strain was replaced with three distinct isolates. It also showed
that intramammary challenge of a subclinically infected quarter with the non-indigenous
strain of S. aureus, in two cows, did not result in permanent replacement of the
indigenous strain with the recently infused non-indigenous strain. In addition to the
original strains A and B, eight distinct strains isolated from the four cows, were
recognised by REFP analysis, indicating that a cow, or even a quarter, may be infected
simultaneously by more than one strain of S. aureus. This study showed that the
examination of multiple colonies per milk sample was required in order to identify the
diversity of S. aureus strains present and to maximise the benefit of bacterial strain
identification as an epidemiological tool in mastitis investigations. As REFP analysis is a
time consuming method, a repetitive extragenic palindromic polymerase chain reaction
(REP-PCR) was developed during this project to facilitate more rapid identification of the
ten S. aureus strains originally identified by REFP.
To investigate the acute phase response as an indicator of inflammation in mastitis,

mammary secretions from the four cows with naturally occurring subclinical mastitis
caused by S. aureus were collected prior to, and following, intramammary challenge with
S. aureus. Serum amyloid A (SAA) was isolated and purified from the serum of bovine
clinical cases by ultracentrifugation, gel electrophoresis and electro-elution. Serum
amyloid A was identified by enzyme linked immunosorbent assay and immunoblotting
using a polyclonal rabbit anti-human SAA antibody. The purified SAA was used to
immunise mice in an attempt to produce a monoclonal antibody specifically against
bovine SAA but this proved to be unsuccessful even when the SAA was bound to a
carrier protein. Increased levels of haptoglobin were identified in mammary secretions
from a single quarter of one cow. This quarter had been originally uninfected and then
challenged by the intramammary route with strain B. In the absence of a polyclonal or
monoclonal antibody to bovine SAA, sodium dodecyl sulphate polyacrylamide gel
electrophoresis was used to identify proteins in mammary secretions with a similar
molecular weight to SAA. A protein of similar molecular weight to SAA was identified in
the same quarter in which the increased haptoglobin levels were measured.
Contents
Abstract ........................................................................ i
List of Contents ........................................................... ii
List of Tables ............................................................... vii
List of Figures ............................................................. iv
Acknowledgements .................................................................................. xi
Declaration ............................................................................................... xii
Page No.
Chapter 1
Introduction ................................................................. 1
1.1 Mastitis ..................................................................................... 1
1.1.1 The D isease........................................................................ 1
1.1.2 Economics ........................................................................ 3
1.1.3 Pathogens ........................................................................ 6
1.1.3.1 Contagious Pathogens............................................... 6
1.1.3.2 Environmental Pathogens.......................................... 8
1.1.3.3 Minor P ath o g en s....................................................... 9
1.1.4 Udder D efen ces................................................................. 10
1.1.4.1 Humoral D efen ces.................................................... 11
1.1.4.2 Cellular D efences....................................................... 12
1.1.4.3 Lymphocytes.............................................................. 13
1.1.4.4 Macrophages.............................................................. 13
1.1.5 Treatment and C ontrol....................................................... 14
1.2 The Staphylococci................................................................... 15
1.2.1 G enus.................................................................................. 15
1.2.2 Pathogenicity ................................................................... 17
1.2.2.1 The Staphylococci ................................................... 17
ii
1.2.2.2 Staphylococcus aureus ............................................. 19
1.2.3 Identification...................................................................... 22
1.2.3.1 Methods ................................................................... 22
1.2.3.2 The Staphylococci .................................................... 24
1.2.3.3 Staphylococcus aureus ............................................. 25
1.3 Inflammation ........................................................................... 26
1.3.1 Acute Phase Response .................................................... 27
1.3.2 Acute Phase Proteins ....................................................... 28
1.3.3 Major Acute Phase Proteins of C a ttle .............................. 31
1.3.3.1 Haptoglobin .............................................................. 31
1.3.3.2 Serum Amyloid A .................................................... 32
1.3.4 Mammary Secretions ....................................................... 35
1.3.4.1 Inflammation ............................................................ 35
1.3.4.2 Acute Phase Response and Acute Phase Proteins. . . 35
1.4 Aims of the Project ................................................................. 36
Chapter 2
Infection in Bovine Staphylococcus aureus Mastitis
2.1 In tro d u c tio n ............................................................................. 38
2.2 Materials and Methods ......................................................... 38
2.2.1 Experimental Animals ....................................................... 38
2.2.2 Intramammary Infusion with Staphylococcus aureus 39
2.2.3 Milk Sample C ollection.................................................... 41
2.2.4 Routine Bacterial Isolation from Milk Sam ples............... 41
2.2.5 Identification of Staphylococcus aureus Strains ............ 41
2.2.5.1 Restriction Enzyme Fragmentation Pattern Analysis 41
2.2.5.2 Repetitive Extragenic Palindromic Polymerase
Chain Reaction ...................................................................... 44
2.3 Results ..................................................................................... 45
2.3.1 Infection in Staphylococcus aureus Mastitis ................. 45
2.3.1.1 Isolation of Staphylococcus aureus from Milk Samples 45
iii
2.3.1.2 Differentiation of Staphylococcus aureus Strains . . . 47
2.3.1.3 Discrimination Between Strains of Bovine
Staphylococcus aureus by Polymerase Chain Reaction .... 56
2.3.2 Dynamics of Intramammary Infection Following
Experimental Challenge with Staphylococcus a u r e u s ...... 62
2.3.2.1 Cows Challenged with the Indigenous
Strain of Staphylococcus aureus .......................................... 62
2.3.2.2 Cows Challenged with the Non-indigenous
Strain of Staphylococcus aureus .......................................... 65
2.4 Discussion ............................................................................... 68
Chapter 3
Inflammation in Bovine Staphylococcus aureus Mastitis
3.1 In tro d u c tio n ..................................................................... 78
3.2 Materials and Methods ......................................................... 79
3.2.1 Identification and Quantification of Acute Phase
Proteins from Sera of Diseased C a ttle ............................... 79
3.2.1.1 Determination of Haptoglobin Concentration .... 79
3.2.1.2 Determination of Serum Amyloid A Concentration . 80
3.2.2 Isolation and Purification of Serum Amyloid A ............ 80
3.2.2.1 Separation of Serum from Blood ........................... 80
3.2.2.2 Removal of Non-Lipoprotein Serum Proteins .... 82
3.2.2.3 Determination of Total Protein Concentration of Serum 82
3.2.2.4 Separation of Serum Amyloid A from High
Density Lipoproteins ............................................................ 83
3.2.2.5 Concentration of Serum Amyloid A ....................... 84
3.2.3 Identification of Serum Amyloid A ................................ 84
3.2.3.1 Enzyme Linked Immunosorbent A ssa y ........... 84
3.2.3.2 Sodium Dodecyl Sulphate Polyacrylamide
Gel Electrophoresis .............................................................. 86
3.2.3.3 Immuno B lo ttin g .............................................. 87
3.2.3.4 Immuno Dot-Blot ................................................... 88
Proteins in Milk Samples from Mastitic Cows ......................... 88
3.2.4.1 Determination of Haptoglobin Concentration .... 88
3.2.4.2 Determination of Serum Amyloid A ....................... 89
3.2.5 Development of a Monoclonal Antibody to Serum
Amyloid A .................................................................................. 90
3.2.5.1 Immunisation of Mice with Bovine Serum Amyloid A 90
3.2.5.2 Immunisation of Mice with Bovine Serum
Amyloid A - Apoferritin Complex ........................................ 91
3.2.5.3 Measurement of Antisera Response to Immunisation
of Mice with Bovine Serum Amyloid A ................................. 92
3.3 Results ..................................................................................... 93
Proteins from Sera of Diseased C a ttle ........................................ 93
3.3.2 Determination of Serum Amyloid A Concentration . . . . 94
3.3.2.1 Isolation and Purification of Serum Amyloid A ... 94
3.3.2.2 Identification of Serum Amyloid A ......................... 94
3.3.3 Development of a Monoclonal Antibody to
Bovine Serum Amyloid A ......................................................... 101
3.3.3.1 Immunisation of Mice with Bovine Serum Amyloid A 106
3.3.3.2 Immunisation of Mice with Bovine Serum
Amyloid A - Apoferritin Complex ...................................... 106
Proteins in Milk Samples from Mastitic Cows ....................... 106
3.3.4.1 Quantification of Total Protein and Haptoglobin 106
3.3.4.2 Detection of Serum Amyloid A ............................ 107
3.4 Discussion .............................................................................. 116
Chapter 4
General Discussion .................................................... 121
V
Glossary
References
List of Tables
Page No.
1.1 Common staphylococcal species and their h osts.................... 16

2.1 The double cross-over design for intramammary challenge with
Staphylococcus a u reu s................................................................... 40
2.2 Routine bacteriological screening from all quarters from cows 125,
186, 520 and 703 .......................................................................... 46
2.3 Strains, unrelated to strain A, strain A* or strain B, recognised by
REFP analysis from cows 125, 186 and 703 and designated Rito R 7 55
2.4 Percentage similarity between strains Ri to R7 and strain A,
strain B or Strain A*........................................................ 57
2.5 Number of matching fragments identified between strains Ri - R 7
and strain A, strain B or Strain A *.................................. 58
2 .6 Staphylococcus aureus strains isolated from each quarter of
cow 125, identified by REFP analysis ........................................ 63
2.7 Staphylococcus aureus strains isolated from each quarter of
2 .8 Staphylococcus aureus strains isolated from each quarter of
2.9 Staphylococcus aureus strains isolated from each quarter of

3.1 Serum amyloid A concentrations in diseased cattle .......... 81
3.2 Serum haptoglobin and serum amyloid A concentrations in diseased
cattle .............................................................................................. 95
3.3 Standard serum amyloid A (cone 192|ig/ml) used for
construction of calibration curves for bovine serum amyloid A . 96
3.4 Standard BSA used for construction of calibration curves for
vii
total protein measured by BCA a s s a y ........................................ 98
3.5 Total protein concentration for each fraction following
ultracentrifugation........................................................................ 100
3.6 Total protein concentration of electro-eluates 1, 2 and 3
using the BCA total protein assay ............................................... 105
3.7 Haptoglobin and total protein concentration in mammary
secretions from the hind quarters of cow 125 ........................... 108
secretions from the hind quarters of cow 186 ........................... 109
secretions from hind quarters of cow 520 ................................... 110
secretions from hind quarters of cow 703 ................................... Ill
viii
List of Figures
Page No.
2.1 REFP of bacterial isolates from cow 186 on day -4 (pre-challenge)

from RH and LH quarters.............................................................. 48
2.2 Graphical output of digitised image from figure 2.1 ............... 49
2.3 REFP of bacterial isolates from cow 520 on day 17 post-challenge
from cow 520 from RF and RH quarters......................................... 50
2.4 Graphical output of digitised image from figure 2.3 ............. 51
2.5 REFP of bacterial isolates from cow 186 on days 3 and 17 post-challenge
from RH q u a rte r............................................................................. 52
2.6 Grapliical output of digitised image from figure 2.5 ............. 53
2.7 Graphical output of digitised image of REFPs (after Hha I
digestion of genomic D N A )............................................................ 54
2.8 PCR amplification products obtained using REP-1 and REP-2
primers after agarose electrophoresis using a 1.2% agarose gel . 59
2.9 PCR amplification products obtained using REP-1 and REP-2
primers after agarose electrophoresis using a 1.2% agarose gel . 60
2.10 Restriction digestion of REP-PCR amplification products
after agarose electrophoresis (1.2% agarose)................................ 61
3.1 Calibration curve for bovine serum amyloid A .................... 97
3.2 Calibration curve for BCA total protein a s s a y .................... 99
3.3 SDS-polyacrylamide gel (15%) electrophoresis of acute phase
bovine serum fractions following ultracentrifugation...... 102
3.4 Immunoblot after SDS-polyacrylamide gel (15%)
electrophoresis of acute phase bovine serum fractions following
ultracentrifugation........................................................................... 103
3.5 Coomassie stained SDS-polyacrylamide gel (15%) electrophoresis
of the HDL portion of acute phase serum following ultracentrifugation,
dialysis and lyophilisation............................................................ 104
of milk samples from cow 703 .................................................... 113
of a control milk sample ‘spiked’ with a range of concentrations of
semi-purified serum amyloid A ..................................................... 114
of a control milk sample, semi-purified serum amyloid A and suspected
serum amyloid A milk sample....................................................... 115
x
Acknowledgements
I am indebted to Dr. Julie Fitzpatrick for her supervision, enthusiastic

encouragement and invaluable guidance during the course of this study, and also
to Dr. David Eckersall for joint-supervision and use of the laboratory in the
Department of Veterinary Clinical Biochemistry. I am also greatly indebted to Dr.
David Platt for his supervision and the use of facilities within the Bacteriology
Department, Glasgow Royal Infirmary.
I wish to express my gratitude to:

Eric Mackay, Rhone Poulenc Diagnostics Limited, for collaboration on the
monoclonal antibody production.
Dr. Jacquie McKeand, Department of Parasitology, for help and guidance with
electro-elution.
Dr. Lubna Nasir, Department of Veterinary Clinical Studies, for her help and
guidance with the REP-PCR.
Dr. David Logue, Scottish Agricultural College, Auchincruive, and Shelagh Pitt,
Catriona Ritchie and Lindsay Eaglesham, for preliminary sampling and
intramammary infusion.
Professor Gruys, Utrecht, Netherlands, for the kind donation of monoclonal
antibodies.
I would also like to thank Karen Logan, Arlene Macrae, Andrew Barbour, Steven
McFarlane and all the other Veterinary Medicine and Biochemistry Laboratory
staff for their help and support throughout this project.
Finally, I would like to acknowledge the University of Glasgow James Houston

Crawford Postgraduate Scholarship for funding this study.
xi
Apart from the help acknowledged I declare that the work described
was carried out by me and is not that of any other person and, further
has not been submitted, in full or in part, for consideration for any
other degree or qualificaton.
Fiona J. Young, May, 1997

Chapter 1
Introduction
1.1 Mastitis
1.1.1 The Disease
Bovine mastitis is a serious welfare and economic problem in the United

Kingdom. Mastitis is recognised as one of the major diseases adversely affecting
dairy cow welfare and is the single most common cause of death in adult cows
(Menzies et al.y 1995). Peracute and acute mastitis are recognised as significant
causes of poor welfare in dairy cows (Logue and Hillerton, 1996), whereas the
major impact of subclinical mastitis is in reduction of farm income (Blosser, 1979;
Rebhun, 1995). Mastitis causes greater financial loss to the dairy industry than
any other disease (De Graves and Fetrow, 1993), with 70% o f this loss
attributable to reduced milk production caused by cases of subclinical mastitis,
many of which remain undetected for a considerable time (Sandholm and Mattila,
1986). Mastitis affects both the quality and quantity of the milk produced (De
Graves and Fetrow, 1993; Blowey and Edmondson, 1995). The reduction in yield
of milk is due to irreversibly damaged milk secretory tissue and the compositional
changes in the milk which occur during a mastitic attack, which adversely affect
the processing qualities of the milk and its products (Castle, 1985). Within the
quarter, a low grade infection may steadily destroy the secretory cells, especially
1
the ducts, which are then replaced by scar tissue, thus reducing the capacity of the
quarter to produce milk (Webster, 1993). Compositional changes in the milk vary
depending on the severity of infection, the causative micro-organism involved, the
stage of lactation of the cow and the susceptibility of the individual animal
(Webster, 1993). Mastitic milk is unacceptable to the consumer and some
mastitis pathogens constitute a public health risk, including some strains of
Staphylococcus aureus which produce enterotoxins known to cause gastro
enteritis in humans (Batish et al., 1988; Harvey and Gilmour, 1990; Orden et al.,
1992).
Mastitis, defined as inflammation of the mammary glands, produces clinically

recognisable signs of heat, pain, swelling and inflammation of the affected
quarters (Blowey, 1990). The majority of cases of bovine mastitis result from
colonisation of the mammary gland by pathogenic bacteria (Sandholm and
Mattila, 1986). Many different bacteria cause mastitis and the diversity of
bacteria initiating infection is one of the major problems in rational treatment of
the disease (Webster, 1993). Mastitis can be classified into peracute, acute,
subacute, chronic and subclinical cases depending on the severity of the attack
(Radostits et al., 1994). In peracute mastitis, the udder becomes grossly inflamed
and the resultant mammary secretions are considerably altered. The cow usually
displays symptoms of severe systemic illness which can be fatal (Webster, 1993).
Acute mastitis is less severe than peracute, however the udder is still swollen and
inflamed and the animals temperature may be elevated, although the systemic
symptoms are less marked than in the peracute form (Black, 1992). Subacute
mastitis is characterised by mild inflammation, including abnormal milk secretion,
whereas chronic mastitis is associated with recurrent attacks of inflammation
(Radostits et al., 1994). The cow with a chronically infected quarter is not only
less productive, but acts as a continual source of infection to other quarters and
cows (Webster, 1993). Subclinical mastitis is difficult to detect, with no visible
abnormality of the milk or udder, and an elevated somatic cell count is taken as
the usual indicator of infection. Somatic cells include epithelial cells, lymphocytes
and monocytes but mainly consist of polymorphonuclear leucocytes (PMNL)
(Lam, 1996). The main factor that determines the somatic cell count (SCC) is the
2
infection status (Shepers et al., 1996) but other factors include the number of
infected quarters, the causative organism, the age of the cow and the stage of
lactation (Emanuelson and Funke, 1991).
1.1.2 Economics
Mastitis is one of the major causes of economic loss to the dairy industry. An
average of 40 cases of mastitis per 100 cows occurs per year in UK herds and an
annual incidence of mastitis of less than 30% in any herd is very rare (Esselmont
and Spincer, 1992; Webster, 1993). Mastitis may present as either clinical or
subclinical disease, or a combination of the two. It is, therefore, difficult to
estimate the economic loss as there is an enormous variation within herds in levels
of clinical and subclinical mastitis, so that a herd may have a high subclinical level
but a low clinical level or vice versa. An estimate of the prevalence of subacute,
acute and peracute mastitis cases based on figures suggested by Blowey (1986) is
70% for subacute cases, 29% for acute cases and 1% for peracute cases. The
cost of an average case of mastitis in 1992 was £168 with a range of £60.46 to
£2248, depending on the severity of the case (Esselmont and Spincer, 1992). An
estimate of the number of dairy cows in the UK in 1991 was 2.6 million (June
1991, Agricultural and Horticultural Census), and calculations based on the
prevalence values above, suggest that mastitis costs the UK dairy industry more
than £100 million each year.
The cost of each episode of clinical mastitis results from: 1). discarded milk, 2).
cost of intramammary antibiotics prescribed by a veterinarian; 3). the subsequent
reduction in the milking potential of the affected quarters and poorer quality of
the milk produced (Blowey and Edmondson, 1995). Antibiotic residues in milk
and dairy products cannot be completely eliminated by pasteurisation or boiling
(International Dairy Federation, 1979) and pharmacological legislation requires
that milk be discarded during treatment and for a specified time period after
intramammary infusion (Vautier and Postigo, 1986). Mastitis leads to a reduction
in yield, in lactose and in butterfat content but milk protein levels increase slightly
3
with mastitis (Shuster et a l, 1991; Blowey et al, 1992). For example, a herd
with a somatic cell count of 750,000 cells/ml could be losing 750-9001 of milk per
cow per year, 50g/kg lactose per cow per year and 30g/kg of milk fat per cow per
year (Blowey et al, 1992). In severe cases of mastitis the costs may include the
replacement cost of the cow due to premature culling or death during the mastitic
episode (Esselmont and Spincer, 1992).
Farmers receive financial penalties or premiums based on the level of somatic cells
in milk and the total bacterial count (TBC) of the milk. Somatic cell counts can
be measured at the level of the bulk tank, the individual cow or the individual
quarter.
Bulk milk somatic cell counts (BMSCC) have become universally adopted as a
screening test for milk quality in herds. They are also useful in creating awareness
of the existence of a mastitis problem in a herd, so that when a BMSCC exceeds
permissible levels, further mastitis investigation using more specific techniques is
often undertaken (Radostits et a l, 1994). The BMSCC is particularly useful as an
indicator of subclinical infection within a herd (Booth, 1985; Blowey et al, 1992).
A BMSCC of 250,000 cells/ml is now considered to be a realistic upper limit for
UK herds with good mastitis control (Blowey et a l, 1992). Field studies in
Scotland show that the main cause of high BMSCC in herds in Scotland is
subclinical mastitis, especially when caused by Streptococcus agalactiae and S.
aureus (Logue et a l, 1995).
The individual cow SCC (ICSCC) is an indirect test based on the number of
somatic cells in a composite sample including milk from all four quarters
(Radostits et a l, 1994) and has been shown to be the most efficient way of
identifying high SCC cows (Blowey and Edmondson, 1995). Individual cow SCC
of less than 250,000 cells/ml are considered to be below the limit indicative of
inflammation while counts of > 400,000/ml probably suggest the presence of
infection.
4
Individual quarter SCC (IQSCC) is the collection and examination of milk from
each individual quarter of the udder. Individual quarter SCC exceeding 240,000
cells/ml are considered to be excessive, although most healthy quarters show less
than 100,000 cells/ml (Blowey and Edmondson, 1995).
Total bacterial counts in the milk are also used as indicators of intramammary
infection. In the milk of acute clinical cases the TBC may be as high as 100
million bacteria/ml, subclinical quarters may contain between 1,000 and 10,000
bacteria/ml and normal quarters usually yield less than 1,000 bacteria/ml. The
bulk tank TBC should be ideally < 10,000 bacteria/ml (Blowey and Edmondson,
1995).
The deregulation of the Milk Marketing Boards in 1994 was the biggest change
within the UK dairy industry for over sixty years. The UK’s five Milk Marketing
Boards provided valuable information on national output in the last decade against
quota; provided technical support, including mastitis control; carried out research
and development on milk production and negotiated the best selling price for milk
with the many milk buyers (Sumner, 1994). Under the Milk Marketing Scheme
there was a barrier between producers and users, and industry management
decisions were made by the milk boards who represented the producers, which
resulted in a reduced producer awareness about the market place. On the 1st
November 1994, the statutory requirement for the Milk Marketing Boards to buy
all the milk produced was removed allowing direct purchasing from the farmers
(Pearce, 1995). This allowed purchasers to influence the quality of the product
and allowed full traceability of the supply to the producer.
The EC hygiene directive was implemented initially in the United Kingdom in May
1995 by the Dairy Products (Hygiene) Regulations 1995 (1). This replaced the
existing milk hygiene legislation, The Milk and Dairies (General) Regulations
1959 (Madders, 1995). The directive sets maximum levels for SCC and TBC in
milk. Initially this standard is being implemented at the level of bulk consignments
to milk buyers however, from the 1st July 1997, the standard will apply at
individual producer level. Supplies with a three month geometric mean SCC
5
higher than 500,000 cells/ml and a raw milk TBC of higher than 100,000 over a
two month geometric mean will become unsaleable. From the 1st January 1998,
the standard will be a maximum of 400,000 cells/ml with a raw milk TBC of less
than 100,000 bacteria/ml (Baines, 1996). The implementation of EC 92/46 will
produce some difficulties for UK milk producers, in particular for the minority of
producers (approximately 10%), who have a BMSCC consistently over the
accepted threshold of 400,000 (Logue et al., 1995).
1.1.3 Pathogens
Organisms that cause mastitis can be divided into three main categories depending
on their pattern of infection, i.e. contagious mastitis, environmental mastitis and
summer mastitis pathogens (Blowey et al., 1992).
Mastitis pathogens can also be subdivided into major and minor categories. The
major pathogens are well-recognised as the cause of clinical or subclinical mastitis
and include organisms such as the contagious pathogen S. aureus and the
environmental pathogen E. coli (Shukken et al., 1989). The role of minor
pathogens is less clear with the possibility that they may either cause mastitis
(Counter, 1981; Shukken et al., 1989) or play a role in protecting the udder
(Shukken et a l, 1989; Nickerson and Boddie, 1994; Pankey et al., 1985; Lam et
al, 1996).
1.1.3.1 Contagious Pathogens
Primary udder pathogens, including S. aureus and S. agalactiae, are considered to

be the major causes of contagious mastitis (Webster, 1993). Some authors
consider S. dysgalactiae to be a contagious pathogen (Hillerton et al, 1995),
while others believe it should be classed as an environmental, and therefore,
opportunistic pathogen (Smith et al., 1985; Pankey et a l , 1987; Oliver et a l,
1993). Primary udder pathogens, which are usually transmitted between quarters
6
during the milking process, are natural inhabitants of the animals skin or udder,
and are the predominant cause of subclinical mastitis (Fox and Gay, 1993).
Streptococcus agalactiae is found only within the udder (Fox and Gay, 1993).
Infection with S. agalactiae causes acute inflammation of the affected quarter
with clots forming in the foremilk, and often results in an elevation of the animal’s
temperature. S. agalactiae infection has remained relatively easy to control due
to the micro-organism’s continued susceptibility to antibiotic treatment
(McKellar, 1996). Subsequent infections caused by S. agalactiae cause less acute
inflammation and pain and thus infections may go unnoticed and continue to
erode the milking capacity of the quarter.
Staphylococcus aureus is a natural inhabitant of the skin and accounts for 25-
30% of bovine mastitis cases (Sutra and Poutrel, 1993). If the organism
penetrates the teat sufficiently to cause infection, two main forms of
staphylococcal mastitis occur (Webster, 1993). The peracute form occurs rarely,
but usually occurs in early lactation when the defences of the mammary gland to
infection are at their lowest. Peracute staphylococcal mastitis is extremely severe
and can often result in the death of the animal within twenty four hours of the
appearance of clinical signs. Chronic and subclinical infections with S. aureus are
the most common result of infection (Sutra and Poutrel, 1993). No obvious
clinical signs of infection or alteration in milk composition occur in subclinical
cases. However, clots in the milk may be present during chronic infection
(Radostits et al, 1994). Staphylococci invade deep into the gland (Blowey,
1990) and can survive and multiply within phagocytes (Sandholm and Mattila,
1986), resulting in the host defence mechanisms and subsequent antibiotic
treatment being incapable of completely eliminating the micro-organism from the
udder. Treatment of staphylococcal mastitis is further complicated by the
existence of many strains of S. aureus which can cause infection, each with
variations in sensitivity to antibiotics (Webster, 1993; McKellar, 1996).
Streptococcus dysgalactiae is most frequently transmitted during the milking

process and is found naturally in the udder and on skin teat lesions (Blowey,
7
1990), however, the method of transmission of S. dysgalactaie remains unclear
(Lam, 1996).
1.1.3.2 Environmental Pathogens
Escherichia coli and Streptococcus uberis are the organisms most frequently
associated with environmental mastitis (Smith et al., 1985). Environmental
pathogens are an extremely heterogeneous group of bacterial genera, species and
strains which are naturally found in the faeces, bedding or on the skin of cows
(Smith et al., 1985).
The coliform, E. coli, is a natural inhabitant of the large intestine and is found
surviving in faecal matter (Blowey, 1990). Escherichia coli mastitis can result in
a wide range of clinical signs varying from mild to peracute cases of mastitis. The
variation in clinical signs is dependent on the dose of organism and also the stage
of lactation of the animal (Webster, 1993). In late lactation, or during the dry
period, only mild infection usually occurs, whereas in early lactation, severe
infection can result in peracute mastitis associated with a high mortality rate. The
peracute mastitis caused by E. coli infection is difficult to distinguish from
peracute contagious mastitis caused by S. aureus (Webster, 1993).
Streptococcus uberis is found naturally on mucous membranes, skin and in the

faeces (Blowey, 1990) and generally causes less severe infection than E. coli.
Streptococcus uberis can be isolated from many anatomical sites of the cow
(Wilson, 1981). A survey of 400 herds in the UK showed that the prevalence of
udder infection caused by S. uberis was approximately 15% (Wilson, 1981).
Streptococcus uberis was a frequent cause of new intramammary infection during
the non-lactating period and under natural conditions, existing infections in the
dry period may act as the source of infection during lactation (Smith et al., 1985).
A large proportion of these infections was cleared by the animal shortly after
parturition, with the organisms remaining sensitive to penicillin when treated
(McKellar, 1996). Streptococcus uberis is now increasing in importance as a
cause of mastitis as the levels of contagious mastitis decline (Pott, 1996).
8
Summer mastitis generally affects dry cows and heifers at pasture, especially in
warm humid weather during the months of July to September (Rebhun, 1995).
Several organisms can be isolated from infected quarters, although the major
agent is Actinomyces pyogenes which can be isolated in 78% of summer mastitis
cases (Blowey, 1990). Other organisms involved in summer mastitis include S.
ciysgalactiae, S. indolicus, and micrococci. Infection is transmitted to the
damaged teat end by females of the head fly Hydrotea irritans (Webster, 1993).
Most cases of summer mastitis are acute and severe with generalised toxaemia,
signs of weakness and fever, and the udder is hard and swollen on palpation
(Webster, 1993). Most cases of summer mastitis result in severe and permanent
destruction of the affected quarter or occasionally death of the animal, while less
severe cases can occur and cause scarring of the sinus and teat canal resulting in a
blockage of secretion of milk when the cow calves (Radostits etal., 1994).
Other environmental organisms which can cause mastitis include Pseudomonas,

Klebsiella, Bacillus, Pasteurella species, Enterococcus faecalis and yeasts
(Radostits et al.t 1994).
1.1.3.3 Minor pathogens
Micro-organisms classed as ‘minor pathogens’ are usually considered to be non-

pathogenic and occur naturally in ‘normal’ udders (Bramley and Dodd, 1984).
Minor pathogens include the coagulase negative Micrococcaceae and
Corynebacterium bovis (Lam, 1996). Minor pathogens have been shown to have
a potentially protective effect against infections with major pathogenic bacteria
(Pankey et al., 1985; Nickerson and Boddie, 1994; Lam et al., 1996) and a higher
prevalence of minor pathogens may be associated with a lower incidence of
clinical mastitis (Schukken et al., 1989). Minor pathogens have also been shown
to be associated with cases of clinical mastitis (Counter, 1981; Schukken et al.,
1989) and induce a limited increase in the somatic cell count (Hogan et al., 1987;
Lam et al., 1996 (b)). While some authors argue that the prevalence of minor
pathogens, especially coagulase negative micrococcaceae should be reduced
9
(Myllys, 1995; Smith and Hogan, 1995), others believe that moderate levels of
somatic cells induced by minor pathogens may play a role in protecting the gland
against infection, particularly against coliform bacteria (Lam et al., 1996).
1.1.4 Udder Defences
The bovine mammary gland has many forms of defence against invasion by
pathogenic organisms, combining non-specific and specific systems, including the
anatomical features of the gland and the humoral and cellular defence mechanisms
(Outteridge and Lee, 1988).
The teat duct provides the initial non-specific defence against bacterial
penetration, acting as both a physical barrier and as a source of antimicrobial
substances (Outteridge and Lee, 1988). The physical barrier is provided by the
smooth muscle sphincter surrounding the teat canal which inhibits bacterial
penetration by maintaining tight closure and limiting bacteria to the teat orifice
(McDonald, 1975). The teat canal is partially occluded by keratin, a mesh-like
substance derived from the teat canal lining which inhibits the progression of
bacteria through the teat duct into the udder (McDonald, 1970). Keratin also
provides the source of antimicrobial substances including lipids and proteins
(McDonald, 1970). The importance of the teat duct keratin in protection against
mastitis was demonstrated by Miller et al. (1991) when removal of 70% of teat
duct keratin significantly reduced the ability of cows to resist mastitis for two days
following its removal. Physical variations in teat canal and teat duct have also
been shown to be important factors in mastitis susceptibility (Lacy-Hubbert and
Hillerton, 1995).
10
1.1.4.1 Humoral Defences
This form of defence can be subdivided into a specific category due to the action
of immunoglobulins and a non-specific category due to the action of cytokines
and other proteins. Immunoglobulins form part of the specific humoral defences
of the mammary gland. The immunoglobulin content of milk and colostrum also
increases during inflammation (Nickerson, 1985). Immunoglobulins play a role in
opsonisation and phagocytosis, toxin neutralisation, prevention of bacterial
adhesion to tissues and direct bacteriolysis. IgGi, IgG2 , IgM and IgA have all
been clearly distinguished in bovine mucosal secretions (Duncan et al., 1972).
IgGi functions predominantly as an agglutinating, virus neutralising and antitoxic
antibody (Colditz and Watson, 1985), but can also act as an opsonin (Nickerson,
1985) and is able to fix complement in vitro (McGuire et al., 1979). IgG2 is
cytophilic for ruminant neutrophils (McGuire et al., 1979) and has been shown to
be particularly important in enhancing phagocytosis of S. aureus in cattle
(Watson, 1976; Guidry et al., 1993). The IgA class of immunoglobulins can
inhibit binding of bacteria to epithelial tissue and may reduce absorption of antigen
through the epithelium (Williams and Gibbons, 1972). IgM binds secretory
component and may be secreted in a similar way to IgA (Butler, 1986). IgM has
also been shown to play a role in opsonisation of bacteria by neutrophils (Williams
and Hill, 1982). Immunoglobulin concentration in the mammary gland varies with
the lactation cycle, the highest concentration occurring near parturition (Concha
et al., 1980) and at the end of lactation (Guidry et al., 1980).
Non-specific defences include lactoferrin, lysozyme and lactoperoxidase and

complement (Reiter, 1978). Cytokines may also be included as non-specific
humoral factors as they can be produced by different types of stimuli, although in
many cases their production may be induced by cells responding to specific
immunological stimuli (Wood etal., 1991).
11
1.1.4.2 Cellular Defences
Non-infected quarters usually have a SCC of 1-3 x 105 cells/ml milk and during
infection this can rise to many millions per ml (Nickerson, 1985). The influx of
bacteria and very high levels of SCC can be measured at 12-24 hours after
infusion (Hill, 1981; Shuster et al., 1996). A certain level of SCC in the gland
may in fact be desirable as it is argued that the presence of cells in the secretion
may in fact be protective against infection (Bramley, 1978). Non-specific
leucocytosis, induced by the introduction of a polythene loop into the mammary
gland, an intramammary device, has been shown to cause a mild chronic
inflammation with maintenance of neutrophils within the gland resulting in a
decreased likelihood of establishment of infection (Paape and Corlett, 1984).
Quarters infected with minor pathogens, bacteria which are generally considered
to be non-pathogenic, show increased resistance to superinfection (Bramley,
1978; Linde et al., 1980) and attempts to intentionally infect quarters with minor
pathogens to provide a stimulus to prevent infection against major pathogens have
been made (Linde et al., 1975; Brooks and Bamum, 1984).
Four cell types are present in normal bovine mammary secretions - neutrophils,
macrophages, lymphocytes and epithelial cells and the proportion of these cell
types varies between individuals and depends on the stage of lactation (Lee et al.,
1980). Phagocytosis of foreign material including bacteria is the primary function
of neutrophils, although they are also involved in many processes including
modulation of vascular permeability (Wedmore and Williams, 1981) and release of
inflammatory mediators (Goetzl, 1980). In comparison to blood neutrophils,
mammary neutrophils are less efficient at phagocytosis. This has been reported
to be due to prior phagocytosis of casein micelles (Russel and Reiter, 1975) and
fat globules (Paape and Guidry, 1977). Despite the factors limiting the
phagocytic activity of mammary neutrophils, this activity is still thought to be the
primary antibacterial defence mechanism of the mammary gland.
12
1.1.4.3 Lymphocytes
The percentage of lymphocytes in cells from mammary secretions are relatively

constant throughout the lactational cycle (Lee et al, 1980). The presence of both
T and B cells in bovine milk was suggested by the work of Smith and Shultz
(1977) as cells isolated from milk were shown to respond to T and B cell
mitogens. Concha et al. (1978) isolated lymphocytes from milk and showed that
the proportion of B and T cells was 20 and 45% respectively, similar proportions
to those found in blood. Smith and Goldman (1968) demonstrated that human
colostral lymphocytes were responsive to both phytohaemmagglutinin (PHA) and
specific antigens. B lymphocytes and plasma cells are present in the mammary
gland tissue and in the teat and play a role in synthesis of immunoglobulins
(Nickerson and Heald, 1982; Collins et al, 1986). T lymphocytes are involved in
cell mediated immune response by producing lymphokines that act as
chemoattractants for neutrophils and macrophages and for stimulating leucocyte
microbial activity (Craven and Williams, 1985).
1.1.4.4 Macrophages
Cells of the monocyte-macrophage series are the major type in mammary

secretions for most of the lactational cycle (Watson and Lee, 1978). They are
also involved in the induction and development of the local immune responses
(Concha and Holmberg, 1990). Mammary macrophages from involved ovine
glands (Outteridge and Lee, 1981), lactating women (Mori and Hayward, 1982)
and cows (Fitzpatrick et al, 1995) have been shown to function as antigen- and
mitogen-presenting cells in lymphocyte stimulation tests.
In spite of the presence of considerable numbers of immune cells in the local

mammary gland environment, it has been suggested that the mammary gland is
generally immunologically compromised when compared to the rest of the body
(Hurley et al., 1990). The activity of all types of white blood cells in milk,
neutrophils, lymphocytes and macrophages, have each been shown to be reduced
compared to cells isolated from blood (Paape et al., 1975). The interaction
13
between cells in the local mammary gland environment is also important in
induction of immunity and failure of cell co-operation is another possible
mechanism whereby udder defences may be impaired.
1.1.5 Treatment and Control
All dairy farmers are encouraged to practice mastitis control, partly to reduce the
wastage that the disease causes, but also to contribute to the maintenance of milk
quality (Radostits et al., 1994). Neave et al. (1969) introduced a standard
mastitis prevention programme for contagious pathogens which has been
continuously revised and developed, and has resulted in an approximately 50%
reduction in udder disease over the past 25 years (Bramley and Dodd, 1984;
Blowey and Edmondson, 1995). Great progress has been made in mastitis control
in the UK during this period, mainly due to the uptake of the ‘Five Point Plan’ by
dairy farmers (Blowey and Edmondson, 1995). The five point plan was outlined
by the National Institute for Research in Dairying and the Central Veterinary
Laboratory in 1969 and includes recommendations to: apply an approved teat
disinfectant after every milking; treat clinical cases of mastitis promptly; infuse
long-acting antibiotic into all quarters at drying off; test milking machines
annually; and cull cows showing repeated cases of clinical mastitis (Blowey et al.,
1990).
The five point plan provides an economic advantage, is within the scope of the
dairyman in terms of technical skill and understanding, can be easily introduced
into the management system employed, and encourages farmers to persist with the
plan by rapidly reducing the occurrence of clinical mastitis and the rejection of
milk by milk processors on the grounds of quality (Blowey et al., 1990).
The use of antibiotic therapy as a control measure for bovine mastitis is not ideal
due to potential public health risks resulting from the use of intramammary
antibiotics and the subsequent risk of antibiotic residues in meat and dairy
products (Vautier and Postigo, 1986). The use of intramammary antibiotics can
14
adversely affect manufacturing processes for cheese and cultured milks (Sandholm
and Mattila, 1986). To enable improved methods of treating and preventing
mastitis to be developed, the basic defence mechanisms and immune responses
within the bovine udder should be fully understood. A greater understanding of
the individual micro-organisms which induce mastitis including strain diversity,
toxin production, and their subsequent effects on the host should be determined to
aid the development of future mastitis control programmes.
1.2 The Staphylococci
1.2.1 Genus
The staphylococci have been studied for over a century, during which time their
pathogenic role in many diseases of man and animals has been established. The
staphylococci are normally associated with the mucous membranes, skin and skin
glands of warm blooded animals and are found frequently as aetiological agents
causing mastitis, food poisoning and different forms of wound and systemic
infections (Makaya, 1996). Currently the genus Staphylococcus is comprised of
at least thirty one distinct species (Kloos and Bannerman, 1994). Some common
staphylococcal species and their natural hosts are listed in table 1.1.
Members of the genus Staphylococcus are facultatively anaerobic, non-motile,

Gram positive cocci which can occur singly, in pairs or irregular clusters (Kloos
and Schleifer, 1986; Prescott et al., 1990; Fox and Gay, 1993). The
staphylococci, like the micrococci, are catalase positive. However, they differ
from the micrococci in several ways: the staphylococci are oxidase negative,
ferment glucose anaerobically, have a cell wall composition of peptidoglycan and
techoic acids, and cell wall DNA with a lower guanidine and cytosine content
(approximately 30-39% compared with at least 60% in the micrococci) (Prescott
eta l., 1990).
15
Species Coagulase Common, Natural host
S. aureus + primates, humans, mammals, birds
S. intermedius + carnivores, mammals, birds
S. epidermidis humans, domestic animals
S. xylosus humans, birds, domestic animals
S. warneri domestic animals, birds,
S. hyicus + swine, cattle, goats
Table 1.1: Common staphylococcal species and their natural hosts
16
The ability of the staphylococci to produce coagulase can be used to divide them
into pathogenic and relatively non-pathogenic strains (Oeding, 1983). Coagulase
positive strains often produce a yellow carotenoid pigment and can cause acute or
chronic infections, whereas strains that do not synthesise coagulase are non-
pigmented and generally less pathogenic (Prescott et al., 1990). The most
commonly known and well described coagulase positive staphylococci are S.
aureus and S. intermedius (Sears et al., 1993). Staphylococcus aureus is one of
the most common staphylococci isolated from a wide range of animal species,
either as normal flora or as an aetiological agent associated with purulent
infections (Matsunaga et al., 1993). The coagulase negative staphylococci make
up the largest group within the genus and consist of more than thirty different
species (Aarestrup, 1995). Non-haemolytic, coagulase negative staphylococci
especially S. epidermidis, S. wameri, S. hyicus and S. xylosus are classified as
minor mammary pathogens (Radostits et al., 1994). Mastitis resulting from these
organisms may be subclinical and result in decreased production of milk and an
elevation in somatic cell count in infected glands (Rebhun, 1995).
1.2.2 Pathogenicity
1.2.2.1 The Staphylococci
A number of factors have been shown to be important in the pathogenicity of

staphylococci, including the route of infection, enzyme or toxin production and
components of the bacterial cell wall. The route of staphylococcal infection is
important in determining the outcome of the invasion of the organism as generally
only small numbers of staphylococci are required to produce infection of the skin
or skin glands such as the sebaceous and mammary glands (Blowey et a l, 1992).
For example, at least 109 colony forming units of some strains of S. aureus are
required to kill mice by intraperitoneal or intravenous inoculation, whereas only
nine colony forming units are required to kill mice by intramammary inoculation
17
(Anderson, 1986). The immediate availability of milk as a good growth medium
as well as the absence of sufficient numbers of phagocytic cells, until they are
recruited by chemotaxis, allows rapid multiplication of the staphylococci within
the mammary gland (Anderson, 1986).
Staphylococci induce infection through their ability to multiply and spread widely
in tissues, which is aided by their production of many extracellular substances
such as exotoxins or enzymes which are thought to be involved in invasiveness
(Kotzin et al, 1993; Lee, 1996). The pathogenicity of the staphylococci is related
mainly to their synthesis of toxins such as haemolysin, leucocidin and enterotoxin,
or of enzymes such as coagulase, hyaluronidase and penicillinase (Anderson,
1986). The function of these extracellular enzymes and toxins should be
considered not only as destructive to the host, but as advantageous to the
invading organism in providing low molecular weight nutrients for bacterial
growth (Anderson, 1986; Makaya 1996). Coagulase, hyaluronidase,
phosphatases, DNAase, proteinases, lipase, catalase, lactate dehydrogenase and
penicillinase are the most frequent enzymes produced by the staphylococci,
however, while not all enzymes are produced by every strain, every strain does
produce catalase (Prescott et al, 1996). Most of the enzymes are produced in the
late logarithmic phase of growth with the exception of coagulase which is
produced during early exponential growth (Arvidson, 1983). The main toxins
produced are leucocidin, exfoliative toxin, toxic shock syndrome toxin (TSST-1),
haemolysins and enterotoxins (Wood et al, 1991; Kenny et al, 1993). The
enterotoxins are membrane-damaging exoproteins which can be divided into
several different types, designated staphylococcal enterotoxins A-E (Wood et al,
1991; Orden et al, 1992). The function of alpha toxin, one of the enterotoxins,
and leucocidin, is directed towards the destruction of cells and lysis of neutrophils
(Anderson, 1986; Prescott et a l, 1990).
Several other substances which are involved in resistance to phagocytic ingestion

are found within the cell wall structure of the staphylococci. Protein A binds
immunoglobulin via the non-specific Fc receptor rather than the antigen specific
Fab terminal and thus activates complement which increases the resistance of the
18
staphylococci to phagocytic ingestion by masking the immunoglobulin Fc sites
(Greenberg et al., 1987; Roitt et al., 1993; Lee, 1996). Peptidoglycan which is
the major component of the staphylococcal cell wall can induce a number of
endotoxin like effects, including pyrexia (Adlam and Easmon, 1983).
1.2.2.2 Staphylococcus aureus
Staphylococcus aureus is a particularly well adapted pathogen for the mammary

gland (Pott, 1996). It frequently causes clinical mastitis, which is mainly a
sporadic disease resulting from chronic subclinical infection.
The pathogenic capacity of a particular strain of S. aureus is due to the combined

effect of extracellular factors and toxins, together with the invasive properties of
the strain (Prescott et a l, 1990). S. aureus produces around thirty extracellular
proteins, some of which have been shown to be virulence factors. Matsunaga et
a l (1993) compared the distribution of virulence factors among S. aureus isolates
from peracute, acute and chronic bovine mastitis cases and correlated the results
with clinical presentation. These authors concluded that all S. aureus isolates
from peracute cases produced TSST-1, staphylococcal enterotoxin C, alpha-
haemolysin and beta-haemolysin, whereas none of the peracute isolates produced
clumping factor or protein A, which were associated with chronic mastitis.
However, there was no obvious variation in virulence factor production between
isolates from acute cases or isolates from chronic cases which is in agreement
with the findings of Jonsson and Holmberg (1981).
The initial line of defence against staphylococcal infection is the physiologically

normal and intact skin. In terms of infection of the udder, the action of the teat
duct sphincter (McDonald, 1975) and the antibacterial properties of the keratin
lining the teat duct play an important role. Mechanisms such as flushing of the
udder during milking also aid in preventing infection (Anderson, 1986).
Attachment to ductal epithelial cells is an important step in colonisation of the
mammary gland and surface epitopes are likely to be involved (Blowey et al,
1992; Makaya, 1996). Antibody has been shown to play a role in preventing
19
adherance of bacteria to mammary gland epithelial cells (Kenny et al, cited in
Blowey et al., 1992). Once the outer skin defences of the host are breached the
staphylococci are killed almost exclusively by the phagocyte-opsonin system
(Anderson, 1986). Many staphylococcal products are directly or indirectly
chemotactic for neutrophils and macrophages (Kenny et a l, 1993). However, in
small quantities, alpha toxin, leucocidin and peptidoglycan may be antichemotactic
(Craven and Williams, 1985). Recruitment of neutrophils is generally not
impaired, and ingestion by phagocytes is not a limiting factor as long as
recruitment of neutrophils is sufficiently fast (Craven and Williams, 1985). The
migration of neutrophils from the blood into the udder, involves the cells passing
through the blood-milk barrier and is activated by chemotaxis (Kremer et al,
1990). Once the neutrophils have passed into the milk in response to chemotaxis,
they engulf the invading micro-organism. Inside the neutrophils, bacteria are
destroyed by a system involving hydrogen peroxide (Blowey and Edmondson,
1995). Staphylococci are killed by neutrophils by discharge of the granular
contents into phagosomes (Lee, 1996). Staphylococcus aureus avoids elimination
from the udder by employing a number of features including the ability to survive
inside white blood cells and to outlive these cells and thus survive within the
udder. Host inactivation of strains of S. aureus is a delicate balance between
staphylococcal growth and the phagocyte-opsonin system, and usually this
balance reaches equilibrium at the level of intracellular survival of the
staphylococci, which is a typical feature of chronic staphylococcal mastitis in
cattle (Anderson, 1982). The inability of phagocytes to kill all intracellular
staphylococci is not fully understood and may relate to the exhaustion of
bactericidal substances, the metabolic dormancy of the organism or the low pH of
the phagocytic vacuole (Rebhun, 1995). The development of methods to kill all
intracellular staphylococci remains one of the most important and urgent problems
in staphylococcal research (Anderson, 1986).
Abscess formation and fibrosis associated with staphylococcal infection reduces

phagocyte infiltration and prevents antibiotic penetration, with multiple micro
abscesses forming throughout the mammary tissue in chronic mastitis (Sears and
Heider, 1981; Makaya, 1996). When ingestion of invading organisms occurs,
20
many staphylococci are killed by the neutrophils, however sufficient levels of
alpha toxin may have been produced to lyse the phagocytes and cause necrosis of
the adjacent tissue (Anderson, 1986).
A major limitation in the effective control of staphylococcal mastitis remains its

poor response to antibiotic therapy. There are a variety of reasons for this of
which antibiotic resistance plays only a minor role (McDonald and Anderson,
1981; Newman et al., 1985; Craven, 1987). The ability of bacteria to survive
inside polymorphonuclear neutrophils, macrophages and epithelial cells, where
they are protected from the action of certain antibiotics, may also contribute to
their resistance to therapy (Sandholm and Mattila, 1986; Blowey et a l, 1992). At
present, staphylococcal infections are treated most successfully with a bactericidal
antibiotic such as the semisynthetic penicillins which have some ability to
penetrate intracellular organisms surrounded by dense tissue (Anderson, 1986).
However, Madgwick et a l (1989) concluded that penetration of certain
antibiotics into bovine neutrophils can be altered by the presence of ingested
staphylococci, and that high intracellular levels of antibiotics does not necessarily
ensure effective intracellular activity against pathogenic micro-organisms.
Vaccination against staphylococcal bovine mastitis has been attempted but has
met with limited success (Slanetz et al., 1963). A S. aureus mastitis vaccine is
commercially available in the United States of America which has been shown to
cause decreased prevalence of S. aureus positive milk cultures and decreased
persistence of intramammary infection following experimental infection (Williams
et al., 1966; Williams et al., 1975). Anderson (1986) suggests that no single
virulence determinant has been identified from which a vaccine against
staphylococci can be developed. Many virulence factors play a role in the
pathogenesis and, therefore, antibodies against them may have only a partial role
in protection. Development of a multi-component vaccine comprising of capsular
polysaccharide complexed with either a toxoided staphylococcal toxin or with a
surface protein known to be important in colonisation of the mammary gland
epithelial cells has been suggested by Foster (1991). The multiplicity of virulence
factors produced by S. aureus in addition to this organisms ability to evade the
21
host immune system by a variety of mechanisms has resulted in S. aureus being a
common mastitis pathogen which is particularly difficult to treat and control.
1.2.3 Identification
1.2.3.1 Methods
Species identification of a microorganism is important for diagnosis, treatment

and prevention of disease. Identification and additional typing can be carried out
on the basis of growth characteristics of an organism on selective media, the
recognition of microbial antigens by monoclonal or polyclonal antibodies,
antibiotic susceptibility, biochemical characteristics, susceptibility to infection by
bacteriophages and general features such as odour and colony formation in
addition to molecular typing techniques (van Belkum, 1994).
In bacterial typing, the reliability of a typing system depends on the bacterial

characteristic on which the typing method is based. The choice of the
characteristic depends on its stability within a strain and its diversity within a
species (Eisenstein, 1990; Brown, 1991). Methods used for discrimination of
strains within a species can generally be divided into phenotypic and genotypic
procedures. The use of conventional phenotypic and molecular genotypic typing
methods as epidemiological tools help in understanding infectious diseases (Kapur
et al., 1995). Control of communicable diseases, such as mastitis, can be
improved by the use of typing methods which define sources of infection,
mechanisms of transmission, and rates of spread within a population (Makaya,
1996).
Phenotypic procedures take advantage of biochemical, physiological and

biological phenomena, whereas genetic procedures aim to detect polymorphisms
at the level of nucleic acids or to detect allelic variation at the level of enzymes
(Kapur et al., 1995). Some of the commonly used conventional phenotypic
22
methods include antibiogram typing (Aarestrup et al., 1995 (c)), biotyping
(Devriese 1984), serotyping and phage typing (Guinee and Van Leeuwen 1978;
Lipman et al., 1996; Makaya, 1996). These biological tests are laborious to
perform and require sophisticated reagents which are often not commercially
available. However, these methods can provide adequate epidemiological
analyses of various organisms. Alternatively, protein analysis can be used to
delineate the origin of a strain of microorganism and to establish relationships
among isolates (Eisenstein, 1990). Techniques such as whole-cell protein
profiling, outer membrane profiling, isozyme electrophoresis and various
immunoblotting techniques are frequently employed (Eisenstein 1990).
Phenotypic procedures are generally not used for discrimination between strains
of different species, whereas genetic procedures are preferred when strain
differentiation is required (van Belkum, 1994).
Genotypic methods are systems which have been developed more recently for use
in bacterial typing. These methods are based on less environmentally influenced
genetic bacterial elements, and as such, they have a higher typability,
reproducibility and discriminating power compared to the phenotypic methods
(Makaya, 1996). Commonly used genotypic methods are plasmid profiling
(Aarestrup etal., 1995 (c)), direct sequencing of DNA fragments, or variations of
restriction fragment length polymorphism techniques such as ribotyping
(Aarestrup et al., 1995 (b)), pulsed field gel electrophoresis, polymerase chain
reaction (PCR) -digest and random amplification of polymorphic DNA fragments
(Wolcott, 1992; Olsen et al., 1993; Matthews et al., 1994; Aarestrup, 1995;
Rantamaki and Muller, 1995; Lam et al., 1996; Lipman et al., 1996). However,
genetic typing assays also have drawbacks. In general these procedures require
relatively large amounts of high quality DNA or RNA and a high degree of
technical skill. Therefore, simple procedures such as PCR, in which small
amounts of relatively impure nucleic acid is required have been developed (van
Belkum, 1994). PCR typing provides the potential to analyse most of the
medically important organisms by a single technique. The bacterial genome
harbours repetitive sequences that can be used for DNA typing (Versalovic et al.,
1993; Lupski and Weinstock, 1992). The characteristics of these sequences are
23
their restricted length and their widespread occurrence. Prokaryotic repetitive
motifs occur throughout the entire genome but rarely within genes. Repeats were
initially discovered in E. coli and Salmonella typhimurium (Higgins et al., 1982;
Gilson et al., 1989), and were later identified in several other bacterial species.
PCR assays aiming at prokaryotic consensus repeats, arbitrary sequences or
sequences of the methicillin resistance gene complex detected nearly twice as
much variation as phage typing, indicating that PCR fingerprinting enabled the
detection of genetic variants in the homogeneous phage groups (van Belkum,
1994).
1.2.3.2 The Staphylococci
DNA restriction endonuclease fingerprinting (REF) measures restriction fragment

length polymorphism and has been developed for many pathogens including the
staphylococci (Bumie and Lee, 1988; Bialkowska-Hobrzanska et al., 1990).
Jayarao et a l (1991) reported that DNA REF was an effective method for
differentiating closely related and unrelated strains of S. uberis isolated from
bovine mammary secretions, which was in agreement with studies by Christ et al.
(1988) and Hill and Leigh (1989). Another study by Matthews et al. (1992)
examined 51 staphylococcal isolates by REFP and antibiogram typing and
concluded that DNA REFP anlaysis successfully differentiated closely related
strains of each species isolated. The ease by which REF analysis can be
performed together with the reproducibility and clarity of REF patterns suggest
that this technique is useful in the differentiation of closely related and unrelated
strains of staphylococci isolated from bovine mammary secretions (Matthews et
a l , 1992).
24
1.2.3.3 Staphylococcus aureus
Many phenotypic and genotypic methods have been used to study S. aureus strain
diversity in cases of bovine mastitis (Fox et al., 1990; Roberson et al., 1994;
Aarestrup e ta l., 1995 (b); Lam et al., 1996). A staphylococcal biotyping system
was developed which differentiated S. aureus strains from man and animals into
host-specific ecovars and biotypes which were not host-specific (Devriese, 1984).
Problems with biotyping include the strict need for adherence to the described
methods in order to obtain reproducible results. Multilocus enzyme
electrophoresis (MLEE) has been used extensively to index allelic diversity among
human and animal pathogens including the staphylococci. (Matthews et al,
1994). Kapur et a l (1995) typed 90% of S. aureus isolates using MLEE, and
clearly demonstrated the presence of genetic heterogeneity among strains isolated
from a single herd. MLEE provides data from which statistical estimates of
genetic diversity and overall chromosomal relationships can be obtained, however,
levels of discrimination are often too low to be of epidemiological value when
used alone.
Phage typing and genotyping by PCR using RAPD, ERIC1R and ERIC primers
was used to differentiate 71 S. aureus strains isolated from bovine mastitis
(Lipman et al., 1996). The RAPD primers gave the most significant DNA
polymorphism changes, with phage typing yielding no more additional
information. ERIC primers were used to compare DNA polymorphism patterns
of different isolates and identified 11 genotypes within 63 S. aureus isolates from
five herds in a study carried out by Lam et al. (1996). Matthews et al. (1994)
also used a PCR based DNA fingerprinting technique to type 75 S. aureus isolates
into 19 distinct profiles and these authors concluded that this technique
differentiated closely related strains within and between geographic locations.
PCR has been shown to be a reliable and practical method for identification of
isolates associated with bovine mastitis although few publications describing the
use o f PCR in S. aureus mastitis are available. The genotype differentiation of S.
25
aureus strains is thought to be one of the most accurate methods to date (Wang et
al., 1993). Compared to existing typing methods, PCR based DNA fingerprinting
is easy to perform and interpret (Matthews et al., 1994), with only Saulnier et al.
(1993) reporting that pulse-field gel electrophoresis was more discriminating.
Most authors conclude that ribotyping is the most discriminatory typing method
followed by biotyping and phage typing which can have a low level of
discrimination due to the low typability of some bovine strains (Aarestrup et a l,
1995 (c, d); Makaya, 1996). Using more than one phenotypic or genotypic typing
method for strain biotyping improves the discriminatory powers. Development of
methods allowing maximum discrimination among S. aureus isolates should help
in further understanding the ecology of these pathogens and the epidemiology of
mastitis caused by them.
1.3 Inflammation
Mastitis is clinically recognisable by pain and inflammation of the udder (Webster,

1993). The heat and redness often associated with mastitis is caused by an
increase in blood flow accompanied by leakage of cells and fluids into the tissues
resulting in increased pain within the infected area (Janeway-Travers, 1994).
Inflammation is a result of dilation of local blood vessels in conjunction with
increased permeability to leucocytes and is a complex response to tissue injury.
The response can be either local or systemic, or a combination of the two (Evans
and Whicher, 1992). The main cell types involved in the inflammatory response
are the polymorphonuclear neutrophilic leucocytes, macrophages, monocytes and
a small proportion of lymphocytes (Gruys e ta l, 1993; Travers, 1994).
Inflammation may be initiated in various ways through different pathways (Evans

and Whicher, 1992). Simple trauma activates mast cells which release mediators
to control vasodilation, vascular leakage and, in some circumstances, proteolytic
26
activation of biochemical mediator systems such as complement (Roitt et al.,
1993). The inflammatory response can be triggered directly by pathogens,
especially in early infection where bacterial pathogens are the main initiating
factors (Vander et al., 1994). During the initial stages of infection, the
inflammatory response is essential in attracting the non-specific inflammatory cells
such as monocytes and neutrophils to the site of infection, where these cells
release inflammatory mediators which activate other components of the immune
response (Bomstein, 1982; Marinkovic et al., 1989; Travers, 1994). The cellular
processes of inflammation fall into four major groups: 1). changes in blood flow
caused by changes in smooth muscle cell function resulting in vasodilation; 2).
alterations in vascular permeability caused by cytoskeletal contraction in
endothelial cells; 3). migration of phagocytic leucocytes to the site of
inflammation; 4). phagocytosis (Roitt etal., 1993).
1.3.1 Acute Phase Response
The acute phase response (APR) in animals is a result of tissue damage and is
particularly associated with inflammation (Eckersall and Conner, 1988). In
addition to the local responses described above, there is also a systemic APR,
which encompasses the animals initial reaction to infection, inflammation or
trauma (Skinner et al., 1991). The term ‘acute phase’ was introduced by Avery et
al. as early as 1941. The APR is a non-specific response which commences
within a few hours of infection, can last for several days and includes pyrexia,
cachexia, and variations in serum hormone, trace element and protein
concentrations (Eckersall, 1992 (b)).
The APR is controlled by a complex array of cytokines and hormones which

produce multiple effects such as fever, lethargy, muscle proteolysis and
leucocytosis (Evans and Whicher, 1992; Eckersall, 1992 (b)). The APR is
initiated by cytokines such as interleukin-1 and interleukin-6 produced by white
blood cells which alter the regulation of proteins secreted by the liver into the
27
bloodstream, with IL-6 acting specifically on hepatocytes (Janeway-Travers,
1994; Vander et al., 1994). The APR is thought to be beneficial to the injured
animal in acting to restore homeostasis and prevent microbial growth (Koj, 1974;
Pepys and Baltz, 1983; Gruys et al., 1994) and has been identified in cattle by
measurement of acute phase proteins (APP) in a variety of clinical conditions such
as salmonellosis, mastitis, acute severe metritis, pneumonia and traumatic
pericarditis (Eckersall, 1992 (a)).
1.3.2 Acute Phase Proteins
During the APR a major alteration is found in the liver where the hepatocytes are
stimulated to increase production and secretion of several plasma proteins
(Vander et al., 1994). These are termed positive acute phase proteins and
increase in concentration in the serum during the APR. The hepatocytes also
regulate the secretion of albumin, a negative APP, which decreases in
concentration during the APR, along with several other negative APP such as
transferrin (Eckersall and Conner, 1988; Gruys et al., 1994). The APP, including
haptoglobin, serum amyloid A (SAA), alpha-1-anti-trypsin, ceruloplasmin and
fibrinogen are mainly glycoproteins secreted by the liver in response to IL-6, IL-1
and TNFa produced by leucocytes and macrophages during infection and
inflammation (Bomstein, 1982; Laurell, 1985; Marinkovic, 1989). Regulation of
synthesis of the APP appears to be a highly complex process which is modulated
or controlled by several cytokines. The major cytokines IL-6, IL-1, TNF-a,
interferon y, leukaemia inhibitor factor and transforming growth factor-p, alone,
or in combination, have been shown to influence the serum concentration of the
acute phase proteins (Maurhaug and Dowton, 1984). Acute phase proteins have
many roles within the body ranging from acting as scavengers, promoting
immunoglobulin production or involvement in tissue repair (Whicher and Dieppe,
1985). As the circulating concentration of these proteins reflects the presence or
extent of infectious or inflammatory lesions they can be used as markers of disease
(Eckersall, 1992 (a)).
28
Acute phase proteins are classified according to the degree of increase in serum
concentration as either major, moderate or minor APP: for example, in humans C-
reactive protein (CRP) is classed as a major APP, as the serum concentration of
CRP increases from a very low level of less than lmg/1 to a concentration of
greater than 200mg/l within twenty four hours (Kushner et al., 1978; Roitt et al.,
1993), whereas a moderate APP in humans, such as alpha-1-anti-trypsin, has only
a 2-3 fold concentration increase within 3-5 days of infection (Eckersall, 1992
(a)). Major APP are useful markers of inflammation if they have negligible or low
basal values with a narrow reference range which remains unchanged with age,
sex, or genetic make-up of the animal. The concentration of a major APP should
increase rapidly to very high levels (>100 times) in response to infection or
inflammatory conditions, with the level of response ideally equivalent to the
amount of tissue damage sustained (Kent, 1992). Non-inflammatory conditions,
nutritional status, exercise, handling or other forms of minor stress should ideally
not affect APP concentration, whereas a secondary infection or relapse in the
inflammatory condition should result in an increase in plasma concentration of an
APP being used to provide diagnostic information (Kent, 1992).
Major APP vary with species, in humans CRP and SAA are major APP
(Pepys,1979), whereas CRP appears to be a normal constituent of bovine serum
(Maudsley, 1985). In general, the APP of the dog and cow are similar to those in
man, however significant differences do exist (Eckersall, 1988). As in man,
canine CRP is a major APP with a normal serum concentration of less than 5mg/l
increasing to over 200mg/l during the APR (Eckersall, 1988). Fibrinogen is
probably the most universally measured APP in animal plasma (Schlam, 1970),
however, it is not a major APP in many animal species, with the most accurate
assay procedures being unsuitable for rapid analysis (Laurell, 1985). Haptoglobin
is a major APP in cattle with a low basal level which increases dramatically during
the APR, yet haptoglobin is a constitutive plasma protein in dogs and humans
(Eckersall, 1988). Levels of haptoglobin have been shown to double in dogs and
humans following surgery (Conner et al., 1988). Haptoglobin and SAA are major
bovine APP (Eckersall and Conner, 1988; Gruys eta l.y 1994).
29
The time course of the APR also varies considerably with the APP studied. In
humans, SAA and CRP increase in concentration very rapidly within hours of the
initial infection, reaching maximum levels in 1-3 days and rapidly resuming basal
levels after the APR has been controlled (Kushner, 1982; Emery and Luqmani,
1993). In contrast, fibrinogen may not reach maximum serum concentration until
two weeks after the initial APR and requires a further seven days to return to
normal levels (Blackburn, 1994). In cattle, SAA has been shown to become
elevated in both spontaneous and induced APR within eight hours of infection
(Alsemgeest et al., 1993).
In addition to using APP levels as markers of disease in the live animal,

measurement of APP may also be useful during meat inspection, for detecting
infected carcasses (Saini and Webert, 1991). Many of the conditions which cause
condemnation of meat are also conditions which would be likely to produce an
APR and thus an elevation of the relevant APP (Saini and Webert, 1991). The
assessment of APP of animals at slaughter could enable detection of the presence
of inflammatory lesions in the carcasses and form an important element of future
quality assurance schemes (Eckersall, 1992 (b)).
30
1.3.3 Major Acute Phase Proteins of Cattle
1.3.3.1 Haptoglobin
Bovine haptoglobin is a highly polymerised serum haemoglobin binding protein

which has been identified as an acute phase reactant in cattle (Morimatsu et al.,
1992). Levels of haptoglobin increase in sera after inflammatory stimuli and has
resulted in haptoglobin being used as an indicator of a variety of inflammatory
conditions (Eckersall, 1988; Morimatsu et al., 1992). The rapid increase in
haptoglobin concentration provides the basis for the usefulness of haptoglobin in
the early detection of the APR (Skinner et al., 1991). Bovine haptoglobin differs
considerably from human haptoglobin in terms of physiology and immuno-
reactivity and is a major APP which is normally absent or present at low
concentration in serum (Eckersall and Conner 1988).
Haptoglobin concentration in bovine serum was originally expressed as a

percentage of haemoglobin binding capacity based on the ability of haptoglobin to
bind haemoglobin and preserve peroxidase activity at low pH. However the
purification of bovine haptoglobin by Eckersall and Conner (1988) has now
allowed the quantification of serum haptoglobin in grams per litre. The use of the
haptoglobin-haemoglobin assay in routine testing of bovine plasma is useful in
monitoring the APR, however, as the function of haptoglobin is to bind and
remove free circulating haemoglobin, the interpretation of results has to be
cautious when a potential haptoglobin response is accompanied by haemolytic
disease, when inaccurate results may be obtained (Eckersall, 1988; Skinner et al.,
1991). In vitro, the formation of the haptoglobin-haemoglobin complex is
followed by the removal of this complex by the liver, resulting in reduced
haptoglobin levels. One advantage of the haptoglobin assay over other APP
assays is that it can be performed quickly and accurately and may, therefore, be
preferable to an assay for other major acute phase proteins, such as SAA, which
have lengthy and technically demanding assay procedures (Skinner et al., 1991).
31
Several other assays for detecting elevation in haptoglobin values have been
developed involving single radial immunodiffusion (Morimatsu et al., 1992) and
high performance liquid chromatography (Salonen et al., 1996). These methods
have also been shown to have improved sensitivity compared to the routine
haptoglobin-haemoglobin assay, but are still affected by removal of haptoglobin in
haemolytic conditions.
In serum of healthy cattle the normal level of haptoglobin is less than 0.10g/l, with
the concentration increasing up to 100-fold within twenty four hours of infection
(Conner et al., 1988). The increase in bovine haptoglobin concentration from a
basal level of less than 0.10g/l to greater than 4.00g/l within 48 hours of infection
may indicate the severity of the physical stress or inflammation in diseases such as
mastitis, pyometra and traumatic reticulitis, and as such, may be a valuable aid in
diagnosis of clinical disease (Makimura and Suzuki, 1982; Skinner et al., 1991).
1.3.3.2 Serum Amyloid A
Serum amyloid A was initially recognised in serum due to its cross-reactivity with
antisera to the amyloid AA protein isolated from deposits of secondary amyloid
(Levin et a l 1973; Husby and Natvig, 1974). Serum amyloid A is a trace level
constituent of normal serum found in the circulation bound to the high density
lipoprotein apoSAA, and is a very sensitive APP which increases up to 1000 fold
after tissue damage and infection in cattle, horses and humans (Kushner, 1982;
Maurhaug, 1983; Alsemgeest e ta l, 1993; Horodagoda, 1994).
Serum amyloid A has been identified as a very heterogeneous protein in all species
studied to date and is regarded as representing a family of 12kDalton molecules
with 104-112 amino acid residues, with a possible insertion of 8 amino acids at
position 72 (Betts et al., 1991). Bovine SAA has 112 amino acid residues and
contains a 9 amino acid insertion between positions 69 and 70 (Rossevatn et al.,
1992). The majority of the SAA isotypes are involved in the APR while others
are constitutive proteins (Whitehead et al., 1992). All types of SAA are
transported in serum as apoproteins in the high density lipoprotein (HDL) fraction
32
(Benditt and Eriksen, 1977). The association of SAA and HDL is likely to occur
following the secretion of the SAA (Hoffman and Benditt, 1982), and in most
species, two structurally similar SAA isotypes are expressed in the liver, while a
structurally different form of SAA is independently regulated and expressed
primarily in the extrahepatic tissues (Maurhaug and Dowton, 1984). The increase
in serum concentration of SAA during inflammation is due to both the increase in
synthesis and decrease in degradation of the SAA by the liver (Maurhaug and
Dowton, 1984).
The function of SAA remains unclear, with original studies by Aldo-Benson and
Benson (1982) indicating that SAA may be influencing lymphocyte responses to
antigens. Other authors have suggested that the function of SAA is to redirect
HDL to inflammatory cells such as macrophages, or to aid in cholesterol removal
(Kisilesky, 1991). The former theory has been supported in studies which showed
that when SAA is bound to the HDL portion, the affinity of HDL for
macrophages increases 2-3 fold. Atherosclerotic lesions possess many features
similar to those found in inflammatory reactions, however the role of SAA in
pathogenesis remains unclear (Maurhaug and Dowton, 1984). Studies by
Shainkin-Kestenbaum et a l (1991) have indicated that a possible feedback system
may exist between SAA and the immunoregulatory cytokines, and as SAA
increases in concentration after tissue destruction, a role in normal tissue repair
may also occur.
Serum amyloid A is the most sensitive APP characterised to date, in both humans
and many animal species, and has been used in the diagnosis and monitoring of
inflammatory and infectious diseases. An increased concentration of SAA has
also been associated with amyloidosis in the dog, mink, cat, rabbit, horse and
cattle (Hoi and Gruys, 1984). Serum amyloid A has been identified as the major
APP in the horse and has been shown to be a reliable marker for assessment of
tissue damage during training and racing (Pepys et al., 1989). Recently SAA has
been identified as an APP in cattle (Boosman et a l, 1990; Alsemgeest et a l,
1993) and monitoring bovine SAA concentration may serve as a specific and
reliable marker for clinical inflammation and tissue injury (Boosman et a l, 1990;
33
Alsemgeest et al., 1993). However, due to the unavailability of a rapid automated
assay, the use of SAA as a routine indicator of inflammation has been limited
(Maurhaug and Dowton, 1984). Measurement of SAA in cattle is hampered by
the difficulty in isolation of the SAA when dissociated from the HDL fraction and
also the poor immunogenicity of the protein for the production of antiserum
(Eckersall, 1992 (a)).
Other problems in the development of a SAA assay have been lack of standards
and monospecific antibodies for SAA (Maurhaug and Dowton, 1984). Improved
methods for purification of SAA and the introduction of monoclonal or polyclonal
antipeptide antibodies have aided in the development of current assays for human
SAA (Casl and Grubb, 1993; Wilkins et al., 1993).
The regulation of SAA and haptoglobin appear to be disease specific pathways.

Serum amyloid A was found to be an indicator of acute inflammatory lesions,
whereas haptoglobin indicated more chronic pathology (Alsemgeest et al., 1994).
In human disease, SAA is a more sensitive indicator of inflammation than
haptoglobin but studies in cattle have been limited due to the non-availability of a
robust SAA assay. In man, the assay of APP has become an accepted and useful
diagnostic aid for the detection and monitoring of conditions leading to an APR
(Kushner and Mackiewicz, 1987). The development of APP detection assays in
the assessment of the chronicity and severity of disease in cattle has considerable
potential for diagnosis and prognosis. Limited studies have shown that the
circulating concentration of bovine SAA correlates with the level of severity of
disease status i.e. acute, subacute or chronic (Gruys et al., 1994; Horadogoda et
al., 1994). In clinical diagnosis in cattle the levels of APP may aid in monitoring
diseases such as hepatic lipidosis, inflammatory processes and periparturient
diseases especially when both haptoglobin and SAA are measured (Gruys et al.,
1993).
34
1.3.4 Mammary secretions
1.3.4.1 Inflammation
Measurement of the SCC is an indirect test used routinely in the analysis of milk
as an indicator of bovine mastitis (Sears and Heider, 1981). Indirect tests which
determine alterations in milk composition, enzyme concentration and serum
protein levels can also be used as indicators of inflammation (Huszenicza and
Stollar, 1993). The NAGase test is based on the measurement of the cell
associated enzyme N-acetyl-(3-D-glucosaminidase and is reputed to be the most
accurate of the indirect tests and as accurate as SCC in predicting the infection
status o f a quarter (Pyorala, 1988). Other indirect tests such as the estimation of
serum albumin concentration, is indicative of damage to the mammary mucosa
and increases according to the epithelial damage sustained (Radostits et al,
1994). The antitrypsin test which measures the trypsin-inhibitor capacity of the
milk has also been extensively used in diagnosis (Radostits et a l, 1994). The
diagnosis of both clinical (Shuster et al., 1991; Huszenicza and Stollar, 1993) and
subclinical mastitis has been carried out using these methods of detection
(Nizamilioglu et al., 1989).
1.3.4.2 Acute Phase Response and Acute Phase Proteins
Limited work has been performed to date on the detection of the APR and APP in
mammary secretions. Schroldt et al. (1995) described the first detection system
for the APP CRP in milk as a novel indicator of bovine mastitis. Milk samples
taken from healthy and mastitic cows gave CRP concentrations ranging from 65-
103ng/ml and 93-417 ng/ml respectively. In some cases, as much as a tenfold
increase in CRP levels was observed in milk following inflammation of the udder,
suggesting the potential usefulness of CRP as an indicator of mastitis. CRP
however is generally not considered as a major APP in bovine or equine serum
(Kent, 1992). Levels of cti anti-trypsin, which is only a moderate serum APP in
35
the bovine, have also been shown to increase in serum and milk of cows with
mastitis (Honkanen-Buzalski et al., 1981; Sandholm et a l, 1984). The
concentration of GCi anti-trypsin was related to the presence of serum albumin and
increased SCC in milk: as the severity of the mastitic attack increased, the levels
of cti anti-trypsin in milk also increased, by 100 fold, while in some animals only a
4 fold increase was detected in serum (Honkanen-Buzalski et al, 1981).
Acute phase protein assays may provide a valuable aid to prognosis and early
diagnosis of conditions such as bovine mastitis (Conner and Eckersall, 1986), and
detection of the major bovine APP, SAA and haptoglobin in milk may also form a
part of future quality assurance schemes. Of these two APP, SAA would be the
better analyte to quantify as it is not effected by haemolytic conditions and is more
responsive to stimulation than is haptoglobin (Horadogoda et al, 1994).
1.4 Aims of the Project
The overall aim of this project was to study infection and inflammation in
experimentally-induced S. aureus mastitis in cows. A novel double-cross over
design was employed for experimental intramammary challenge in this study.
Udder quarters, with naturally-occurring subclinical mastitis caused by S. aureus
were challenged by the intramammary route with very high numbers of either the
indigenous or non-indigenous strain of S. aureus.
In studying infection due to S. aureus, the main aim of this part of the project was
to study the dynamics of intramammary challenge in cows and to determine if
infusion with large numbers of different S. aureus strains induced clearance,
alteration or persistence of the original subclinical mastitis. Another aim was to
determine whether variations in the strain of S. aureus were evident when more
than one colony of S. aureus per milk sample was examined employing recent
molecular techniques such as Restriction Enzyme Fragmentation Pattern (REFP)
Analysis and Repetitive Extragenic Palindromic Polymerase Chain Reaction
(REP-PCR).
36
In studying inflammation due to S aureus, the main aim of this part of the project
was to investigate if SAA could be used as an indicator of inflammation in S.
aureus mastitis and thus as a potential adjunct to future quality control
programmes for dairy products. To facilitate this, the aim was to develop a
monoclonal antibody raised specifically against bovine SAA by developing
methods allowing improved purification of SAA.
37
Chapter 2
Infection in Bovine Staphylococcus aureus

Mastitis
2.1 Introduction
One of the main aims of this part of the project was to investigate the effect of
infusing S. aureus by the intramammary route into quarters already subclinically
infected with S. aureus. It was considered necessary to infuse very large numbers of
S. aureus bacteria into the quarters in order to provide sufficient challenge to the
existing organisms colonising the tissue, and greater numbers of S. aureus were
infused in this study than had been previously described in the literature.
To improve investigation of intramammary infection caused by S. aureus, it was

decided to examine multiple colonies isolated from infected quarters and to use
genetic fingerprinting to identify strain variation among cows and quarters.
Due to the time taken to perform the genetic fingerprinting using standard techniques,
an attempt to develop a repetitive extragenic palindromic polymerase chain reaction
(REP-PCR), which allowed more rapid identification of bovine S. aureus strains, was
made.
2.2 Materials and Methods
2.2.1 Experimental Animals
To investigate the dynamics of intramammary infection with different strains of S.

aureus, four adult Friesian cows with naturally occurring subclinical mastitis caused
38
by S. aureus were studied. The cows were initially selected as having high ICSCC (>
600,000/ml), and the infected quarters were subsequently identified by IQSCC (>
400,000/ml), and S. aureus was isolated on routine bacteriology from them. Two
cows, both from the same farm, were shown to be infected with the same strain on the
basis of genomic DNA fingerprints (described in section 2.2.5.1) and the strain was
designated strain A. Two other cows, both from a different farm, were shown to be
infected by the same method with a different strain, designated strain B.
2.2.2 Intramammary Infusion with Staphylococcus aureus
Each cow described above was challenged with either 108 S. aureus per quarter in the
case of the non-infected quarter, or 109 S. aureus in the case of infected quarters in a
double crossover design, such that one cow from each farm was challenged with the
indigenous strain and one cow from each farm challenged with the non-indigenous
strain (Table 2.1). Throughout the trial all four quarters were used and given the
prefixes LF, RF, LH and RH referring to the left fore, right fore, left hind and right
hind quarters respectively.
Staphylococcus aureus cultures containing 1.25xl09 cells per ml in Brain Heart

Infusion Broth (Oxoid, Basingstoke, Hampshire) were incubated overnight at 37°C
and then centrifuged at 5670g for 10 minutes. The supernatant was decanted, the
pellet was resuspended in 15ml Ringers solution and centrifugation and resuspension
of the pellet was repeated. One and a half millilitres of cell suspension at 1.25xl010
cells/ml was added to 20 ml Ringers solution and infused into infected quarters (i.e. a
total of 109 bacteria), uninfected quarters had 20ml of a 1:10 dilution of the original
cell suspension (i.e. a total of 108 bacteria) infused into them. Control quarters had
only 20ml Ringers solution infused into them. The teats were prepared by washing
and drying prior to thorough disinfection with surgical spirit. The first few draws of
milk were discarded from each quarter immediately prior to intramammary infusion.
The S. aureus cultures were infused using a 6FG dog urinary catheter inserted
through the teat canal to which a 20ml syringe containing the bacterial culture was
attached. Following infusion, the base of the quarter was massaged in an upward
39
Cow Quarters infected Pre challenge Quarters Quarters Challenge
Number prior to trial strain isolated challenged with challenged with strain
1 1 10 * S. aureus 1 1 1 0 1 S. aureus
186 LF, RF, LH, RH A LF, RF, LH, RH B
125 LF, RF, LH A LF, LH A
520 RH B RH RF A
703 LF B LF LH B
Table 2.1: The double cross-over design for intramammary challenge with
Staphylococcus aureus
40
motion while holding the teat duct closed, to ensure thorough dispersion of the
infused material into the gland cistern. None of the infusion was lost via the teat canal
and all teats were then dipped in an iodophor teat dip. The cows were examined
clinically at regular intervals for 48 hours after the infusion procedure and were
milked twice daily by machine.
2.2.3 Milk Sample Collection
Individual quarter milk samples were collected by hand stripping on day -4 (pre
challenge), and on days 3, 8 and 17 post-challenge.
2.2.4 Routine Bacterial Isolation from Milk Samples
Three millilitres of milk was mixed with 17ml of sterile distilled water and centrifuged
at 4500g for 20 minutes. The cell pellet was resuspended in lOOpl of lauryl broth,
from which 50pl was spread onto Columbia blood agar and mannitol salt agar plates
which were incubated overnight at 37°C. After incubation, 15-20 individual S. aureus
colonies, identified by colony morphology, were subcultured individually onto
Columbia agar slopes and incubated overnight at 37°C. Inoculates Columbia agar
slopes were then stored at 4°C.
Further strain characterisation was employed and used restriction enzyme

fragmentation pattern (REFP) analysis.
2.2.5 Identification of Staphylococcus aureus Strains
2.2.5.1 Restriction Enzyme Fragmentation Pattern Analysis
Subculture of Staphylococcus aureus Strains. Ten millilitres of BHI broth (Oxoid)

was inoculated aseptically from Columbia agar culture slopes and incubated overnight
at 37°C. Columbia blood agar streak plates were also prepared and incubated to
confirm the purity of the initial culture.
41
Isolation of Genomic DNA
Purification of DNA. Purification of DNA was carried out over a period of two
days. On day one of the purification procedure, overnight BHI broths were
centrifuged at 4480g for 10 minutes and resuspended in 3ml Tris ethylene diamine
tetra acetic acid sodium chloride buffer (TES) (50mM Tris-base, 50mM sodium
chloride, 5mM disodium EDTA, (pH8.0)). The samples were then aliquoted into 3
eppendorf tubes and microcentrifuged at 13,800g for 30 seconds before being
resuspended in 200pl TES which contained 50mM sucrose to prevent cell lysis.
Twenty microlitres of the endopeptidase lysostaphin (Sigma, lOOOunits/ml) which is
specific for the pentaglycine bridge involved uniquely in the cell wall structure of S.
aureus, and lOOjul lysozyme (Sigma, 40mg/ml) were added and mixed thoroughly
before being incubated at 37°C for 30 minutes. After incubation 20% SDS (15pl) was
added and mixed by inversion, fifty microlitres of Proteinase K (lOmg/ml) was also
added before the sample was sheared through a 25 gauge needle to reduce the
viscosity of the DNA and incubated for two hours at 37°C. One hundred microlitres
of TEio (lOmM Tris base, lOmM disodium EDTA, (pH7.8)) was added to reduce the
DNA concentration before the initial purification stage using phenol chloroform
(500|lU). After mixing thoroughly, the samples were microcentrifuged at 13,800g for
10 minutes and the aqueous layer was transferred to another eppendorf and 500pl of
isopropanol was added. The samples were left at room temperature for 60 minutes to
precipitate the DNA, followed by centrifugation at 13,800g for 10 minutes. The
pellet was resuspended in 100pi TEio and triplicate tubes were pooled, lOOpl of
ammonium acetate (7.5M, pH8.0) and 600pl of 95% ethanol were added and samples
were stored at -20°C overnight, to precipitate the DNA.
On day two, further purification procedures were carried out. Samples were
centrifuged at 13,800g for 10 minutes, resuspended in 300pl TEio and 20pl RNAase
(Sigma, lOmg/ml) was added. After incubation at 37°C for 60 minutes, the
phenol/chloroform, isopropanol and ethanol steps were repeated as above and the
samples were again stored at - 2 0 overnight.
42
Restriction Digestion of DNA. Restriction digestion was carried out on day three,
the day immediately following the purification procedures. Samples were centrifuged
at 13,800g for 10 minutes and resuspended in 60p.l TE (lOmM Tris base, ImM
disodium EDTA (pH8.0)). Restriction digestion of each sample using restriction
enzyme Hha I (Gibco Life Technologies Limited, Paisley) was carried out, with
bacteriophage X DNA digested with Hha I, Kpn I and Pst I as controls. Restriction
products mixed with loading buffer (25g sucrose, 60mg sodium acetate, lOOmg SDS
and 50mg bromophenol blue in 100ml) were electrophoresed in a 0.8% agarose gel
(2:1 dilution with TBE:distilled water, (Tris base 89mM, boric acid 89mM disodium
EDTA 1.25mM, pH8.0-8.4)) at 25mA overnight and the gel was stained using
ethidium bromide in TBE (0.5-1.Oug/ml ethidium bromide). Following removal of
excess background staining by rinsing the gel in distilled water for approximately 15
minutes, the gels were photographed on a transilluminator at 302nm using Polaroid
film type 655.
Digitisation of Restriction Enzyme Fragmentation Pattern Results.

Interpretation of electrophoretic gels was a skill-intensive task. Digitisation of gel
photographs was an objective yet flexible technique that allowed the accurate
assessment of electrophoretic data. The statistical analysis was used to assess the
accuracy of calibration data, and by limiting the percentage fragment size variation,
accurate between-gel comparisons were made. Data was stored in numerical form
(molecular weight of restriction fragments). The graphical output was used to present
the data in a familiar format which allowed visual comparisons with the original gel
photograph. Digitisation of REFP isolates allowed gel to gel-comparisons of S.
aureus strains. Strains were analysed using a commercially available programme
(Platt and Sullivan, 1992) with restriction digested X DNA using enzymes Kpn I and
Pst I as calibration controls. Restriction fingerprints were compared using a
coefficient of similarity calculated from the formula:
SD(%)= _ 2 m_ X100
a+b
43
where ‘m’ was the number of fragments common to two fingerprints and ‘a’ and ‘b’
were the total number of fragments generated from each fingerprint respectively after
digestion by the same restriction enzyme (Dice, 1945).
The molecular weight of the control fragments was fitted to a robust modified
hyperbola, from which the size of restriction fragments in adjacent tracks was
estimated. Numerical values were stored for subsequent graphical output which was
on a logarithmic scale, the experimental variation in fragment size did not exceed 5%
(Plikaytis eta l., 1986).
2.2.5.2 Repetitive Extragenic Palindromic Polymerase Chain Reaction
The repetitive extragenic palindromic polymerase chain reaction (REP-PCR) was used
to obtain a rapid method for the identification of strain diversity in the bovine
staphylococci. Genomic DNA was prepared by isolation of genomic DNA as
described previously in section 2.2.5.1 following the purification stages carried out on
day one and day two during REFP analysis. Fifty nanograms of genomic DNA was
amplified in a total volume of 50pl containing 50pmols of primer REP 1R (5’ 111
NCgNCgNCATCNggC3’, R and D Systems Europe limited, Abingdon) and REP 2
Dt (5’ NCgNCTTATCNggCCTAC3’, R and D Systems), 250pM of dATP, dTTP,
dCTP, and dGTP (Pharmacia, St. Albans, Herts.), lOmM Tris-HCL, 1.5mmol/l MgCl,
50mM/l KCL, 0.1 mg/ml gelatin and 2.5 units of Dynazyme (Flowgen, Lichfield,
Staffordshire). DNA samples were subjected to an initial denaturation of 94°C for 7
minutes, followed by 32 cycles of 94°C for 1 minute, 37°C for 1 minute, 72°C for 8
minutes followed by a final extension of 72°C for 16 minutes using a Perkin Elmer
480 Thermocycler (Perkin Elmer, Warrington). Amplification efficiency was
determined by agarose electrophoresis using a . % agarose gel and viewed as
1 2
described for REFP analysis. Initially, strain A, strain B and strain A* were used for
the optimisation of the protocol, which involved reducing the primer annealing
temperature by 3°C to 37°C, varying the target DNA concentration, primer and
magnesium chloride concentrations.
44
Restriction Digestion of REP-PCR Products. Restriction digestion of the REP-
PCR products was carried out to determine if any further strain pattern variation
could be identified using this method. The digestion enzymes used were Rsa I, Alu I,
Stu I, Hind III, Pst I, Hae III, Eco RI, Bam HI, Hpa I and Kpn I. Each digest
contained lOpl of PCR product, 2pi of appropriate enzyme buffer, 3 pi of enzyme and
distilled water to give a total volume of 20pl. The samples were incubated overnight
at 37°C in a water bath. Following incubation the samples were electrophoresed on a
. % agarose gel, containing ethidium bromide.
1 2
2.3 Results
2.3.1 Infection in Staphylococcus aureus Mastitis
2.3.1.1 Isolation of Staphylococcus aureus from Milk Samples
Routine bacteriological analysis was carried out in the commercial laboratory at the
Scottish Agricultural College, Auchincruive. Routine bacteriological analysis
identified bacteria from each quarter of each cow on each sampling day. Quarters
were recorded as either S. aureus positive or S. aureus negative and the presence of
other bacterial species was also recorded. Initial identification of S. aureus was based
on colony morphology and a positive coagulase test. The distribution of positive
samples is shown in Table 2.2. Throughout the study, uninfected quarters were
designated as control quarters and were analysed to confirm that cross-contamination
had not occurred during the infusion procedure and subsequent sampling.
45
Cow Number Sample Day S. aureus S. aureus
Positive Negative
125 -1 RF, LF, LH RH
125 3 RF, LF, LH RH
125 8 RF, LF, LH RH
125 17 LF, LH RF, RH
186 -1 RF, LF, RH, LH
186 3 RF, LF, RH, LH
186 8 RF, LF, RH, LH
186 17 RF, LF, RH, LH
520 -1 RH RF, LF, LH
520 3 RF, RH LF, LH
520 8 RH RF, LF, LH
520 17 RF, RH LF, LH
703 0 LF RF, RH, LH
703 3 LF, LH RF, RH
703 8 LH RF, RH, LH
703 17 RF, LF, RH, LH
Table 2 .2 : Routine bacteriological results from all quarters from cows 125, 186 and
520 on four sampling days i.e. day -1 (pre-challenge) and days 3, 8 and 17 post
challenge. Routine bacteriological screening from all quarters from cow 703 on four
sampling days i.e. days 0 (pre-challenge), 3, 8 , and 17 post-challenge.
46
2.3.1.2 Differentiation of Staphylococcus aureus Strains
Restriction Enzyme Fragmentation Pattern Analysis and Digitisation of Results.

Bacterial isolates were fingerprinted and analysed by REFP from all four quarters of
each cow on four occasions during the study i.e. day -4 (pre-challenge) and days 3, 8 ,
and 17 post-challenge. A total of 400 cultures were studied by REFP analysis. The
four cows studied were challenged by the intramammary route as shown in table 2 . 1 .
Following digestion of bacterial isolates with restriction digest Hha I, a strain with 25
fragments ranging from 2.92 kilobases to 9.59 kilobases in size was identified and was
designated strain A (figure 2.1, Lanes 6-14). Digitisation of the REFP results is
shown in figure 2.2, Lanes 6-14. A molecular variant of strain A (A*) was also
recognised; 25 fragments were evident following digestion with restriction digest Hha.
I. Twenty-four of the 25 fragments matched those of strain A (figure 2.1, Lanes 1-5).
Digitisation of the REFP results is shown in figure 2.2, Lanes 1-5. The only
difference between strain A and A* was in the largest fragment, with strain A
fragment size being 9.59 kb and strain A* fragment size being 9.32 kb.
Following digestion of bacterial isolates with restriction digest Hha. I, a strain with 30
fragments ranging in size from 2.92-9.59 kb was identified and was designated strain
B (figure 2.3, Lanes 3, 4, 6 , 8-15). Digitisation of the REFP results is shown in figure
2.4, Lanes 3, 4, 6 , 8-15.
Seven genetically unrelated strains were recognised by REFP analysis as strains other
than strain A, strain A* or strain B and were designated strains Ri to R 7 (figure 2.7,
Lanes 5-11). As an example of an unrelated strain, Ri, is shown in lane 8 in figure 2.5
and following digitisation in figure 2.6. The REFP results from additional strains
were digitised (figure 2.7, lanes 5-11) and the total number of fragments identified and
the size of each determined. These additional strains ranged in fragment number from
20-36 fragments with sizes of 2.86-15.88 kb and were isolated both pre and post
challenge (Table 2.3).
47
1 2 3 4 5 6 7 8 9 10 11 1 2 1 3 14 15 16 17
Figure 2.1: REFP of bacterial isolates from cow 186 on day -4 (pre-challenge) from
RH and LH quarters; ethidium bromide (0.5-1.0 fig/ml) stained 0.8% agarose gel.
DNA was digested using restriction endonuclease Hha I. Lanes 1-5 show isolates
from RH identified as A*. Lanes 6-14 show isolates from LH identified as strain A.
Bacteriophage X DNA controls were digested with Hha I, Kpn I and Pst I and are
shown in Lanes 15, 16 and 17 respectively.
48
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Figure 2.2: Graphical output of digitised image from figure 2.2. Lanes 1 -5 show
isolates from RH identified as A*. Lanes 6-14 show isolates from LH identified as
strain A. Bacteriophage X DNA controls were digested with Hha I, Kpn I and Pst I.
Lanes 15 and 16 show Kpn I and Pst I respectively.
49
1 2 3 4 5 6 7 8 9101112131415161718
Figure 2.3: REFP of bacterial isolates from cow 520 on day 17 post-challenge from RF
and RH quarters after digestion with Hha I; ethidium bromide (0.5-1.0pg/ml) stained
0.8% agarose gel electrophoresis. Lanes 1-7 show isolates from the LH quarter.
Lanes 1, 2, 5 and 7 show strain A whereas lanes 3, 4 and 6 show strain B. Lanes 8-15
show isolates from the RH quarter and were strain B. Bacteriophage X DNA controls
were digested with Kpn I, Pst I and Hha I and are shown in lanes 16-18 respectively.
50
51
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Figure 2.4: Graphical output of digitised image from figure 2.3. Lanes 1-7 show
isolates from the LH quarter. Lanes 1, 2, 5 and 7 show strain A whereas lanes 3, 4 and
6 show strain B. Bacteriophage X DNA controls were digested with Kpn I, Pst I and
Hha I. Lanes 15 and 16 show Kpn I and Pst I respectively.
51
1 2 3 4 5 6 7 8 9101112131415161718
Figure 2.5: REFP of bacterial isolates from cow 186 on days 3 and 17 post-challenge
from RH quarter; ethidium bromide (0.5-1.0pg/ml) stained 0.8% agarose gel
electrophoresis. Samples were digested using restriction endonuclease Hha I.
Bacteriophage X DNA controls were digested with Kpn I, Pst I and Hha I and are
shown in lanes 1, 2 and 3 respectively. Lanes 4-10 show isolates from RH quarter on
day 17 post-challenge. Lanes 4, 6, 9 and 10 show strain A*, whereas lanes 5 and 7
show strain A. Lane 8 shows strain Ri. Lanes 11-18 show isolates from RH day 3
post-challenge and were identified as strain A*.
w
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Figure 2.6: Graphical output of digitised image from figure 2.5. Bacteriophage X
DNA controls were digested with Kpn I, Pst I and Hha I. Lanes 1 and 2 show Pst I
and Kpn I respectively. Lanes 3-9 show isolates from RH quarter on day 17 post
challenge. Lanes 3, 5, 8 and 9 show strain A*, whereas lanes 4 and 6 show strain A.
Lane 7 shows strain Ri. Lanes 10-17 show isolates from RH day 3 post-challenge and
were identified as strain A*.
53
M
a;
N
•H
m
+j
c
<u
B
tn
cd
[j-i
1 2 3 4 5 6 7 8 9 10 11
Figure 2.7: Graphical output of digitised image of REFPs after Hha I digestion of
genomic DNA. Strain A (Lane 1), strain A* (Lane 2) and Strain B (Lane 3).
Additional distinct isolates R 1-R7 are shown in lanes 5-11 respectively.
54
00
00 CN
vn ON VO ON
o ON
CN 00I ON 00 o
l i
CO CN On
00 VO
00
00
co
ON on On
CN CN CN CN CN co
OX)
w
CN CN O
M) CN CN CN CN
Q.
00 00 00 00
NO CO CO
00 o o
55
Coefficient of Similarity between Strains. The similarity coefficient between strain
A and strain A* was 96% (24 of the 25 fragments matching). A similarity coefficient
of 83.6% was shown between strain A and strain B with 23 of the 30 fragments
matching. Strain A* and strain B also had a percentage similarity coefficient of
83.6% with 23 of the 30 fragments matching. Strain A and strain A* showed
between 48.9-73.7% similarity to strains Ri to R 7 , whereas strain B had between
52.0-83.9% similarity to strains Ri to R7 (Tables 2.4 and 2.5).
2.3.1.3 Discrimination Between Strains of Bovine Staphylococcus aureus by

Polymerase Chain Reaction
REP-PCR amplification using REP-1 and REP-2 primers on strain A genomic DNA
generated two products of approximately 1,300 base pairs and 2,000 base pairs. The
same method using strain B genomic DNA as the template, generated products of
1,100 base pairs and 2,000 base pairs (figure 2.8). The two strains were, therefore,
distinguished by this method. However, using this method, no difference in the size
amplification of products was observed between strain A and strain A* and identical
banding patterns were seen (figure 2 .8 ).
Under the same REP-PCR conditions and cycling parameters, genomic DNA from
strains Ri to R 7 gave entirely different banding patterns to strains A, B or strain A*
(Figure 2.9).
Restriction Digestion of REP-PCR amplification products. Restriction

endonuclease digestion was carried out to identify any variation between REFP
fingerprinted genomic DNA from strain A and strain A*. The digestion enzymes Rsa
I, Alu I, Hind III, Pst I, Hae III, Eco RI, Bam HI and Kpn I all gave no variation in
amplification product between strain A and strain A*. Only restriction endonuclease
Stu I gave a slight variation in amplification products between strain A and strain A*
(Figure 2.10, Lanes 6 and 7). The variation was the presence of an additional band in
strain A in lane 6 . Digestion enzymes Rsa I, Alu I, Hind III, Pst I and Hae III are
shown as examples of the absence of variation in restriction digestion of REP-PCR
amplification products in figure 2.10 (lanes 2-5, 8-13).
56
Isolate Cow No. Sample Q uarter % Similarity with Strain % Similarity with Strain % Similarity with Strain
Day A B A*
r»
pp
186 RH 63.8 61.5 63.8
t-
125 LF 72.4 66.7 72.4
L89
00
P? e?
703 LF 69.2 68.1
00
57
703 LF 73.7 83.9 73.7
125
P? P?
LH 71.2 75.0 74.6
a
1 00
703 LF 53.3 52.0 53.3
£
00
703 LF 48.9 64.0 48.9
H
H
o
c
(S
o
V
•G
*3
.S
G
G
G
G
G
*3
-t->
O
(A
X)
G
G
G
G
G
<s
G
G
G
G
*G
J3
.2
<u
+-»
4>
•3
i- i
<U
T3
tG
C*h
<L>
T3
>>
-(->
oUh
C/3
X>>
00
T3
CL
<D
.SP
esS
•pM
D £
£
L <
C
Uh o
o
§
V
O
T<3U c-
T3 oo
CL B
PQ
O
(N
§ <43
> £<u
C/J *G
O
c
a>
a>
£
a>
vn JO
J= ■C cn g
o
ed
CO
CN ’§ )
<N ’3
bQ
G
o
§
TD
U
C/3
^ “3
3 m
G O
C"
Cm T3
C
"G
G
S3 ed
C/3 vo
*C/3 00
f-H
00 VO
CN
m VO
"3 P\
CN CN
c
CN
W> A 1"^
Eh CA
M
£
3Ph o
o
>-3
g
T3 o
a> eta
?S C/3
<D
vn c Jd
CN D
u~> rs 3
C/3
t—1
r~
P^ * ’
<
CN CN CN
P? G
C/3
_c ■aM
OX) *3 ■CM
u-* /3
-4
C/3 U
O
G
o PQ
<u
.g
£
■M
<u
•aM
X> t«
C/3 Uh
G O
U
6 <cf
op G
cd
tfcj •a
u.
oo
a >> G
00 00 oo 00 3
O
eS
g
(4-1
o
M
(U
X5
g
3
G
0)
JG
H
vn
CN
3
cd
H
58
12 3 4
Figure 2.8: PCR amplification products obtained using REP-1 and REP-2 primers after
agarose electrophoresis using a 1.2% agarose gel. Lane 1 shows 100 base pair ladder.
Lane 2 shows strain A PCR amplification products of 1,300 and 2,000 base pairs.
Lane 3 shows strain A* amplification products of 1,300 and 2,000 base pairs. Lane 4
shows strain B amplification products of 1,100 and 2,000 base pairs.
59
1 2 3 4 5
Figure 2.9: PCR amplification products obtained using REP-1 and REP-2 primers after
agarose electrophoresis using a 1.2% agarose gel. Lanes 1-5 show amplification
products obtained from additional isolates Ri to R7, previously identified by REFP
analysis.
60
i i n # — **
1 2 3 4 5 6 7 8 9 1 0 11 12 13
Figure 2.10:Restriction digestion of REP-PCR amplification products after agarose

electrophoresis (1.2% agarose). Lane 1 shows 100 base pair ladder. Lane 2 shows
strain A digested with Rsa I. Lane 3 shows strain A* digested with Rsa I. Lane 4
shows strain A digested with Alu I. Lane 5 shows strain A* digested with Alu I. Lane
6 shows strain A digested with Stu I. Lane 7 shows strain A* digested with Stu I.
Lane 8 shows strain A digested with Hind III. Lane 9 shows strain A* digested with
Hind III. Lane 10 shows strain A digested with Pst I. Lane 11 shows strain A*
digested with Pst I. Lane 12 shows strain A digested with Hae III. Lane 13 shows
strain A* digested with Hae III.
61
2.3.2 Dynamics of Intramammary Infection Following Experimental
Challenge with Staphylococcus aureus
2.3.2.1 Cows Challenged with the Indigenous Strain of Staphylococcus aureus
Multiple colonies from each S. aureus positive quarter were analysed by REFP on
each sampling day (tables 2.6 and 2.7)
Cow 125. Cow 125 was identified originally as naturally infected with strain A in the
LF, LH and RF quarters and was challenged with the indigenous strain A into the LF
and LH quarters. In the LH on day -4 (pre-infusion) isolates R 5 and R 2 were
identified although the LH quarter had been identified as having been naturally
infected with strain A on initial screening prior to this sample date. The challenge
strain A was infused into the LF and LH quarters and strain A was isolated in these
two quarters on subsequent sampling days. Strain A was identified in the LH quarter
on day 3 post-challenge indicating that it has successfully colonised this quarter,
remaining until day 17 post-challenge. It is possible that infusion o f large numbers of
strain A into the LH quarter may have replaced the R 5 and R 2 strains as the persisting
organisms. Alternatively, failure to isolate strains R 5 and R 2 may have been simply
due to a dilution effect where the presence of large numbers o f strain A made random
sampling of other strains less likely. The RF quarter which was naturally infected with
strain A and did not receive any bacterial infusion was still infected with strain A at
the end o f the experiment. In the LF and RF quarters on day 17 post-challenge, while
strain A predominated, unrelated strains R 2 (LF) and R 5 (RF) were also isolated as
single colonies.
Cow 703. The LF quarter o f cow 703 was naturally infected with strain B and she
was subsequently challenged with strain B in the LF and LH quarters. Strain B was
isolated on day 3 post-challenge in the LF quarter. On day 8 , strains other than strain
A or B were isolated in the LF quarter and were designated strains R3, R4, R 6 and R7.
Strain B was isolated as a homologous population on day 3 and day 8 from the LH
quarter (Table 2.7). On day nine post-challenge cow 703 suddenly developed signs of
62
Staphylococcus aureus strains isolated from each quarter of cow 125
SAMPLE DAY LF RF LH RH
Infected A Infected A Infected A Control
Challenged A Challenged A
NT
Day -4 (pre AAAAAAAAA AAAAAAAAA r5
challenge) A r2
(9) (10) (2)

NT NT NT
Day 3 (post AAAAAAAAA
challenge) A
(10)
NT NT NT NT
Day 8 (post
challenge)
NT
Day 17 (post AAAAAAA AAAAAAAAA AAAAAAAAA
challenge) r2 Rs
(10) (9)
(8)
Table 2.6: Staphylococcus aureus strains identified by REFP analysis, from four
quarters on four sampling days i.e. day -4 (pre-challenge) and days 3, 8, and 17 post
challenge. Total numbers of colonies subcultured are shown in brackets. A = Strain A,
B = Strain B, A* = Variant Strain A. R2 and R5 = additional isolate. NT = not
analysed. LF, LH, RF, and RH denotes the left fore, left hind, right fore and right
hind quarters respectively.
63
Staphylococcus aureus s tr a in s is o la te d fr o m e a c h q u a r te r o f c o w 7 0 3
Infected B Control Challenged B Control
Challenged B
NT NT NT
Day -4 (pre BBBBBBBBB

challenge)
(9)
NT NT
Day 3 (post BBB BBBBBBBBB

challenge)
(9)
(3)
NT NT
Day 8 (post R 3 R3 BBBBBBBBBB

challenge) R4
R6
R7R7R7R7R7
(8 )
(10)
NT NT NT NT
Day 17 (post
challenge
challenge. Total number of colonies sub-cultured are shown in brackets. A = strain A,
B = strain B. R = additional strain. NT= not analysed. LF, RF, LH and RH denotes
the left fore, right fore, left hind and right hind quarters respectively.
64
severe mastitis including reddening and swelling of the quarters. The cow was
pyrexic and anorexic. Under Home Office Regulations the cow was treated with
flunixin meglumine (Finadyne, Schering-Plough Limited), a non-steroidal anti
inflammatory and an antibiotic amoxycillin/clavulanic acid (Synulox, Pfizer Limited).
No samples were therefore obtained from day 17 post-challenge for cow 703 as
alteration in milk composition prevented the isolation of bacteria for subculture and
REFP analysis.
2.3.2.2 Cows Challenged with the Non-Indigenous Strain of Staphylococcus

aureus
Multiple colonies from each S. aureus positive quarter were analysed by REFP on
each sampling day (table 2.8 and 2.9)
Cow 186. All four quarters of this cow were thought to be naturally infected with
strain A on initial screening, however, while the other three quarters were shown to
be infected with strain A, the RH quarter was shown to be naturally infected with
strain A*. Following challenge of all four quarters with the non-indigenous strain B,
REFP analysis showed that strain B was not isolated from any quarter post-challenge.
Strain A was isolated from the LF, RF and LH in all the sampling days post-challenge
(days 3, 8 and 17). Strain A* was isolated from the RH quarter on day -4 (pre
challenge), and on day 3 (post-challenge) as a homogeneous population. However,
on day 17 post-challenge the RH quarter contained a mixed population of strain A,
strain A* and one other isolate which was not strain A, strain A*, R2 or R5 (identified
in cow 125 previously) and was designated strain Ri (Table 2.8).
Cow 520. Cow 520 was naturally infected with strain B in a single quarter, the RH,
and was challenged with strain A in both the RH and RF quarters. Strain B was
confirmed as the only isolate from the RH quarter pre-challenge (day -4). Strain A
was identified as a homogenous population on all isolates tested on day 3 post
challenge from the RH quarter, but by day 8 post-challenge, strain B was again
identified in all isolates tested from the RH quarter and again on day 17 post-
65
Staphylococcus aureus strains isolated from each quarter o f cow 186
Infected A Infected A Infected A Infected A
Challenged B Challenged B Challenged B Challenged B
Day -4 (pre AAAAAAAA AAAAAAAAA AAAAAAAAA A*A*A*A*A*

challenge) AAA A*A*A*A*A*
(12) (10)
(8) (9)
Day 3 (post AAAAAAAAA AAAAAAAAA AAAAAAAAA A*A*A*A*A*

challenge) AAAAAA AAAAAAAA A*A*A*A*A*
A*A*A*A*A*
A*A*
(17) (17)
(14) (9)
NT NT NT
Day 8 (post AAAAAAAAA
challenge) AAAAAAAA
(17)
NT
Day 17 (post AAAAAAAAA AAAAAAA AA
challenge) AAAAAAAA A*A*A*A*
Ri
(17) (7) (7)
quarters on four sampling days i.e. day -4 (pre-challenge and days 3, 8, and 17 post
challenge. Total numbers of colonies sub-cultured are shown in brackets. A = strain
A, B = strain B and A* = variant strain A. R = additional isolates. NT = not analysed.
LF, RF, LH and RH denotes the left fore, right fore, left hind and right hind quarters
respectively.
66
Staphylococcus aureus strains isolated from each quarter of cow 520
Control Challenged A Control Infected B
Challenged A
NT NT NT
Day -4 (pre BBBBBBBBBB
challenge) BBBBBBBBBB
(20)
NT NT
Day 3 (post AAA AAAAAAAAA
challenge) AAAAA
(3) (14)
NT NT
Day 8 (post AAAAAAAAA BBB
challenge)
(9) (3)
NT NT
Day 17 (post AAAA BBBBBBBBBB
challenge) BBB B
(7) (11)
Table 2.9: Staphylococcus aureus strains identified after REFP analysis, from four
challenge. Total numbers of colonies sub-cultured are shown in brackets. A = strain
A, B = strain B. NT = not analysed. LF, RF, LH and RH denotes the left fore, right
fore, left hind and right hind quarters respectively.
67
challenge. In the RF quarter, strain A was isolated on days 3 and 8 post-challenge.
This result indicates that challenge with strain A resulted in persistence of this strain in
this quarter for the duration of the study. However by day 17 post-challenge, both
strain A and strain B were isolated from the RF quarter (table 2.9).
2.4 Discussion
The accurate determination of intramammary infection status is essential in

conducting mastitis research and herd control (Sears et al., 1990). There are a variety
of diagnostic tests available for identifying mastitis and detecting infection levels in
dairy herds. Methods for diagnosis have classically been divided into two major
categories 1.) the detection of alteration in milk as a consequence of pathophysiologic
changes due to an inflammatory response, 2.) the direct detection of microorganisms
in milk by culture (Sears and Heider, 1981). However, the continued high incidence
of chronic mastitis indicates that current control measures are sub-optimal and that
the development of effective methods will require a more precise understanding of
both the epidemiology of chronic mastitis within and between herds, and also a clearer
view of the host-bacteria relationship within an immunological framework.
In recent years, new bacteriological and epidemiological techniques have become

available to study patterns of infection in bovine mastitis (Lam, 1996). With modem
techniques such as DNA fingerprinting and amplification of DNA by PCR,
information for both diagnosis and epidemiology of intramammary infections may be
provided quickly and reliably to allow detailed studies of natural bovine mastitis
infections (Matthews et at, 1994).
The present study followed the development of a strategic approach to REFP analysis
based on restriction enzymes that recognise a 4-base rather than 6-base cleavage site
in DNA. This technique provides a large amount of information, amenable to
computer analysis and adaptable to a wide range of micro-organisms (Platt et al,
68
1994). Firstly, from case control studies of human S. aureus infection, variation in 1-
3 restriction fragments constituted clonal variation and that unrelated strains differed
in > 7 fragments. Secondly, in studies of bovine S. aureus in Scottish herds there was
evidence of clonal variation within individual herds. This observation was based on
the analysis of a single colony from each sample tested as in the standard diagnostic
procedure. This raised the important question of whether strain variation reflected the
choice of colony for study, genotypic diversification within the herd, or whether
individual cows harboured one or more variants. Furthermore, the chronicity of
bovine S. aureus mastitis raises the possibility that a cow may acquire different strains
of S. aureus at different times that remain co-resident within the udder and would
have consequences for the host-parasite relationship immunologically. Thus one of
the initial aims of this study was to determine whether strain variation was evident
within an individual quarter or cow and to what extent when multiple colonies were
analysed from the same milk sample. Another aim of this project was to study the
dynamics of intramammary challenge with S. aureus in four cows with subclinical S.
aureus mastitis in one or more quarters. A double cross-over design was employed
such that one cow from each source was challenged with the indigenous strain and the
other cow was challenged with the non-indigenous strain. This approach allowed us
to determine if the indigenous strain was displaced by the non-indigenous strain, or
whether superinfection with the indigenous strain altered or had no effect on the
original infection status of that quarter. This study also used a novel approach for
intramammary infusion with S. aureus. Two concentrations of S. aureus were used
for intramammary challenge depending on the original infection status of the quarter:
l.OxlO8 c.f.u/ml. were infused into non-infected quarters, whereas l.OxlO9 c.f.u/ml.
were infused into infected quarters. The high concentration of bacteria infused into
the mammary gland in this study was in contrast to a study by Sears et a l (1990),
which observed the shedding pattern of S. aureus from bovine intramammary
infections. In that study a challenge dose which ranged from 20-2,000 c.f.u/ml was
infused by the intramammary route (Sears et al, 1990). However in the cited study
the strain used was a particularly virulent S. aureus and of 21 previously uninfected
glands challenged with S. aureus, 19 became infected. In the present study, the
strains selected were isolated from cows with subclinical mastitis which are
presumably less virulent than strains involved in clinical mastitis. It was considered
69
that it was necessary to infuse large numbers of bacteria into an infected gland in this
study in order to provide a significant challenge to the existing organisms.
Although bacteriological culture of milk is commonly utilised for diagnosis of

intramammary infection (Sears et al., 1991), the milk collection technique,
transportation and culture methodology should be taken into account when
interpreting the microbial growth from a milk sample. Routine bacteriological
screening generally involves single colony analysis. Many authors including Aarestrup
et al. (1995(a)), Lam et al. (1996) and Birgersson etal. (1992) based their findings on
the single colony analysis of an ‘apparently pure culture’. These apparently pure
cultures were identified by basic methods such as colony morphology, Gram-stain,
coagulase, catalase test, lack of oxidase activity, antibiotic sensitivity and antibiogram
typing. These methods of identification may however fail to identify all strains of S.
aureus especially if they occur at low frequency.
In the present study, routine bacteriological investigation in the commercial laboratory

resulted in S. aureus being isolated from the challenged quarters in three of the four
cows (125, 186 and 520) whether the quarters were previously infected or uninfected
on the majority of occasions. However on three separate occasions (cow 125, day
17, RF; cow 520, day 8 , RF; cow 703, day 8 , LF) bacteriological screening identified
these individual quarters as S. aureus negative. These quarters were subsequently
identified as S. aureus positive following REFP analysis and strain identification in the
research laboratory, which may reflect differences in primary isolation techniques
between the two laboratories. The failure to identify several infected quarters as S.
aureus positive following routine bacteriological screening confirms that routine
bacteriology is not 100 % effective in the identification of intramammary infection.
The study by Sears et a l (1990) concluded that routine bacteriological screening from
a single quarter milk sample was only 75% sensitive in the diagnosis of intramammary
infection caused by S. aureus, thus 25% of infected cows may have a ‘false’ negative
single culture result. Routine bacteriological investigation did not result in S. aureus
isolation from both the challenged quarters (LF, LH) of the fourth cow (703) on all
70
the sampling dates. Cow 703 was naturally infected in the LF quarter with strain B
and subsequently challenged in this quarter with strain B. The LH quarter which was
originally uninfected was also challenged with strain B. S. aureus was not isolated in
any quarters from day 10 post-challenge although it was detected on days 3 (LF and
LH quarters) and 8 post-challenge (LH only). Failure to isolate S. aureus from this
cow from day 10 post-challenge onwards was probably the result of systemic
antibiotic therapy with amoxycillin/clavulanic acid (Synulox) which was administered
on day 9 post-challenge. Failure to isolate the bacteria following therapy probably
indicates that the antibiotic was either effective in killing the S. aureus in the
mammary gland or reduced the shedding of the organism into the milk for the period
of the study.
Restriction enzyme fragmentation pattern analysis was used to determine strain

variation of S. aureus genotypically. REFP was shown to discriminate between both
similar and unrelated strains of S. aureus in this study. REFP allowed discrimination
of two similar strains of S. aureus, strain A and strain A* in the cows. Strain A and
strain A* had almost identical REFP patterns which differed by a single restriction
fragment and were clearly clonal variants. The largest fragment of strain A* was
slightly smaller (9.32 kilobases) than strain A (9.59 kilobases). The two main strains
involved in this study, strain A and strain B, were unrelated. Following digestion of
strain A, 25 fragments ranging from 2.92-9.59 kilobases in size were identified. Strain
B had 30 fragments ranging in size from 2.92-9.59 kilobases. Seven additional
unrelated strains were identified following REFP and designated strains Ri to R 7 .
These additional strains ranged in fragment number from 20-36 with fragment sizes
ranging from 2.86-15.88 kilobases. These findings were similar to those of Jayarao
et al. (1991) who used REFP analysis to identify and differentiate closely related and
unrelated strains of S. uberis isolated from bovine mammary secretions. Jayarao et al.
(1991) analysed 42 strains of S. uberis from 17 cows throughout lactation and
identified 17 REF patterns. Analysis of these isolates by bacteriocin-like inhibitory

substance (BLIS) patterns resulted in only 12 patterns being identified (Jayarao et al.,
1991). Hill and Leigh (1989) also used DNA fingerprinting to analyse S. uberis
isolates and concluded that DNA fingerprinting was a simple and reproducible typing
71
system based on the restriction fragment size of chromosomal DNA. They reported
that the endonuclease Hind III was the most successful in the differentiation of S.
uberis isolates. Matthews et a l (1992) used DNA REFP to analyse 51
staphylococcal isolates from the mammary secretions of cows with subclinical mastitis
and identified 7 S. aureus isolates with 2 distinct REF patterns. Isolates were
evaluated using a computer integrated scanning laser densitometer which allowed the
determination of fragment sizes. Differences in REF patterns were identified as
variation in the number and size of fragments, with each REFP designated as a distinct
pattern. Visual comparison of REFP photographs was also undertaken to verify
results obtained by densitometric scanning. In the present study, strain A* was shown
to be 96% similar to strain A after digitisation and 24 of the 25 fragments matched.
Strain A and strain B showed 83.6% similarity and 23 of the 30 fragments matched.
Matthews et al. (1992), Hill and Leigh (1989) and Jarayao et a l (1991) all used
densitometric scans and graphic displays of restriction patterns in the analysis of
results. Forty-two REF patterns were compared to each other by Jarayao et a l
(1991) and this resulted in the identification of related and unrelated strains. These
methods allowed comparison of strains by fragment size and fragment numbers,
distinguishing similar and unrelated strains on this basis. The accurate comparison of
different strains over a period of time could also be accomplished using this method of
data retrieval and storage.
In this study, a cow with two almost identical strains of S. aureus, which differed by
only one fragment following REFP analysis was recognised. Cow 186 was thought to
be naturally infected with strain A in all four quarters on initial screening. On the
basis of REFP analysis, the RH quarter was shown to be naturally infected with strain
A*. It seems likely that the strain A* was derived originally from strain A. Multiple
colony analysis in this project showed that, on several occasions, more than one strain
of S. aureus was isolated from a single quarter. Both related and unrelated strains of
S. aureus were isolated from the same quarter on the same sampling day in an
individual animal. One example of the isolation of two closely related strains from the
72
same quarter on the same sampling day was the identification of strain A and strain
A* in the RH quarter of cow 186 on day 17 post-challenge. The identification of
unrelated strains of S. aureus from the same quarter on the same sampling day
occurred in other cows in the study. The RF quarter of cow 520 on day 17 post
challenge was shown to contain a ‘mixed’ population of strain A and strain B, the two
main strains in this study, following REFP analysis. Another example of the
identification of two unrelated strains from the same quarter on the same sampling day
was in the LF quarter of cow 125 on day 17 post-challenge when strain A and strain
R 2 and in the RF quarter of the same cow on the same day post-challenge when strain
A and R 5 were identified following REFP analysis. These findings can be compared
to a study by Jayarao et a l (1991) which analysed the genotypic and phenotypic
variation of S. uberis isolates from bovine mammary secretions. Jayarao et a l (1991)
concluded that related strains of S. uberis were isolated in different quarters of the
same gland on the same sample day, isolated in the same quarter of the same gland
over a period of time and also that related strains were detected in different quarter of
the same gland on the same day. Unrelated strains were identified by Jayarao et a l
(1991) from different quarters of the same gland on the same day and also from the
same quarter of the gland over a period of time. Therefore in the present study,
multiple colony analysis by REFP resulted in a more accurate identification of S.
aureus strain diversity within a quarter than single colony analysis by routine
bacteriological methods. Multiple colony analysis is, therefore, a useful technique in
studying bovine mastitis caused by S. aureus.
DNA fingerprinting is another technique used in studying the epidemiology of clinical

and subclinical mastitis (Lipman et a l, 1995; 1996). By comparing isolates between
and within herds, it is possible to draw conclusions about the source and spread of
bacteria. For example, a study by Lam et a l (1996) showed that different E. coli
genotypes were identified from different quarters, suggesting that E. coli is an
environmental pathogen and does not spread from quarter to quarter. In contrast,
Lam et a l (1996) showed that a limited number of S. aureus genotypes (11) were
identified among 63 S. aureus isolates from five herds, supporting the theory that
contagious pathogens transfer between cows and quarters within a herd. S. aureus
73
strain identification using the DNA RAPD assay also showed that the majority of
recurrent cases of S. aureus mastitis were the same genotype, suggesting either
intermittent shedding from a chronically infected quarter or repeated infection with
the same pathogen (Lam et al, 1996). Results from the present study, indicate the
existence of several strains of S. aureus identified from the four cows studied. REFP
analysis identified eight additional strains other than the two main strains in the study,
strain A and strain B, and demonstrated that mixed infection occurs and that the
investigation of a single colony may not provide a complete understanding of infection
patterns.
One disadvantage of REFP analysis is the lengthy process, extending to 2-3 days,
required to obtain bacterial DNA of sufficient purity for subsequent digestion with
restriction endonucleases. This isolation method is similar to the technique used by
Matthews et a l (1992) who used restriction endonuclease fingerprinting of
staphylococcal species of bovine origin. Other methods such as ribotyping performed
by Aarestrup et al. (1995(c)) and MLEE by Kapur et a l (1995) which have been used
in the identification of staphylococci are also lengthy processes. In this study REP-
PCR was developed as a rapid method for the identification of strain diversity in the
bovine S. aureus. REP-PCR was shown to clearly distinguish unrelated strains of S.
aureus whereas closely related strains were not distinguished using REP-PCR. This
technique was shown to discriminate between strain A, and strain B and also among
strains Ri - R7. REP-PCR however, did not discriminate between strain A and strain
A* and was therefore less sensitive than REFP.
Similar techniques have been employed to study variation in other mastitis pathogens.
Jayarao et al. (1992) compared REF and DNA amplification fingerprinting (DAF) in
the subtyping of S. uberis in cattle. REF grouped 22 strains into 12 distinct patterns
whereas DAF grouped them into 15 distinct patterns, indicating that DAF is an
effective typing method and more discriminatory then REF in the typing of S. uberis
strains. Jayarao et al. (1991) suggested that possible variation between typing
methods (REF and BLIS) could be due to the recent identification of S. parauberis
74
which is biochemically, physiologically and serologically identical to S. uberis, with
only a slight genotypic variation differentiating the two strains.
Lam etal. (1996), Lipman etal. (1996) and Matthews et al. (1994) all used variations
of the PCR to differentiate strains of S. aureus isolated from the mammary glands of
cows. Lam et al. (1996) used ERIC primers for differentiation of E. coli and an
arbitrary primer for S. aureus strain differentiation (RAPD assay). Lipman et al.
(1996) used RAPD, ERIC1R and ERIC primers in the differentiation of S. aureus
isolates, identifying and subtyping 71 S. aureus strains. The different primer sets used
by Lipman et al. (1996) gave no variation in the ability to differentiate between strains
under the conditions specified. PCR was shown to be a preferable method to
biotyping as it is a much more specific typing method van Belkum et al. (1993),
Lipman et al. (1995) and Wang et al. (1993). Matthews et al. (1994) used a synthetic
oligonucleotide primer which provided a high degree of resolution in the
differentiation of S. aureus strains with seventy-five S. aureus isolates grouped into
19 distinct profiles.
In the present study, restriction endonuclease digestion of REP-PCR amplification

products was carried out to try to differentiate between strain A and strain A* which
were not discriminated by REP-PCR alone. Of a number of restriction endonucleases
tried, only Stu I gave a variation in amplification products between strain A and strain
A* but this approach in conjunction with standard REP-PCR for the other strains
enabled discrimination of all strains identified from the cows in this project.
The double-crossover design of intramammary infusion with S. aureus allowed study

of the effects of challenge of an infected quarter with the indigenous or non-
indigenous strain of S. aureus. Two cows were challenged with the indigenous strain
of Staphylococcus aureus.
Cow 125 was naturally infected with strain A (LF, LH) and subsequently challenged
with the indigenous strain A in the LF, and LH quarters. The intramammary infection
persisted throughout the study in quarters which were previously subclinically infected
75
or uninfected. This implied that challenge with a high dose of the indigenous strain
did not restimulate the immune system which might have resulted in subsequent
clearance of the infection within the quarter. The identification of additional isolates
both pre-challenge (LH) and post-challenge (day 17, LF: R2; day 17, RF: R5) indicates
the presence of more than one strain within an individual quarter. However, in this
cow, the predominant strain isolated post-challenge remained strain A.
Cow 703 was naturally infected with strain B (LF) and subsequently challenged with
the indigenous strain in the LF and LH quarters. Strain B was isolated post-challenge
from the LH quarter (days 3 and 8) and LF quarter (day 3), however, by day 8 post
challenge, additional isolates R3, R4, R* and R7 were identified in the LF quarter.
Failure to isolate strain B on day 8 post-challenge may be have resulted from an
enhanced immune response to the original bacterial strain B eliminating or reducing
the numbers of this strain, thus allowing different strains to become the predominant
isolates from this quarter.
Two other cows were challenged with the non-indigenous strain of S. aureus. Cow
186, was naturally infected with strain A in three of the four quarters and strain A* in
one quarter and was challenged with the non-indigenous strain B in all four quarters.
In the quarter with the original infection caused by strain A*, by day 17 post
challenge a mixed population of strain A, strain A* and strain Ri was detected. Strain
B was not isolated at any point post-challenge throughout the study.
Cow 520, which was naturally infected with strain B and subsequently challenged
with the non-indigenous strain A, had the non-indigenous strain isolated as a
homogenous population on day 3 post-challenge when all isolates tested were shown
to be strain A. This suggests a transient period of infection by the challenge strain A.
By day 8 post-challenge, only the indigenous strain was identified in all isolates and
this was also the case on day 17 post-challenge. The presence of strain B may be a
result of cross-contamination of strains between cows or quarters. The cows were
tethered separately to reduce environmental spread and the milking machine clusters
76
were disinfected between cows. It is possible that transfer from the RH quarter to the
RF quarter may have occurred during milking in this cow.
This part of the study showed that when infected with the indigenous strain (either
strain A or strain B) and subsequently challenged with the non-indigenous strain, the
indigenous strain persisted as the dominant strain within the quarter and was the only
strain identified on day 17 post-challenge in most quarters in both cases.
77
Chapter 3
3 Inflammation in Staphylococcus aureus

Mastitis
3.1 Introduction
One of the main aims of this part of the project was to develop a monoclonal
antibody raised specifically against bovine SAA. In the initial stages of the
project, SAA concentration in bovine serum samples was assessed by ELISA
using a polyclonal rabbit anti-human SAA antibody. In order to purify SAA from
bovine serum an attempt was made to identify specific diseases of cattle which
might be associated with increased level of this protein. The cattle referred to
Glasgow University Veterinary School included those affected by a variety of
infectious, inflammatory or neoplastic conditions. The aim was to measure serum
haptoglobin levels which had been shown to correlate with SAA levels in diseased
cattle and to screen sera from a number of these cases using the polyclonal rabbit
anti-human SAA antibody. In order to prepare antigen for monoclonal antibody
production, improved methods for isolation and purification of SAA had to be
developed during this project.
78
3.2 Materials and Methods
3.2.1 Identification and Quantification of Acute Phase Proteins

from Sera of Diseased Cattle
3.2.1.1 Determination of Haptoglobin Concentration
Haptoglobin is one of several proteins associated with the acute phase response
and is used as an indicator of inflammation and infection. Increased haptoglobin
levels were detectable in serum within hours of experimental infection of calves
with Pasteurella haemolytica and were correlated with elevated SAA levels
(Horadogoda, 1994). The haptoglobin assay was, therefore, used in this project
to rapidly identify diseased cattle likely to have a high serum amyloid A
concentration.
Haptoglobin Assay. The haptoglobin assay was a semi-automated method based

on that of Conner et al. (1988). Stock haemoglobin solution was diluted in 0.9%
saline to a standard 30mg/ml. All samples, standards and controls were diluted
1:75 in standard haemoglobin solution and left for 10 minutes at room
temperature to allow the formation of the haptoglobin-haemoglobin complex.
Two hundred microlitres of assay reagent (3,3’,5,5’-tetramethylbenzidine (TMB)
stock of 6mg/ml in DMSO diluted 1:100 in chromogen buffer, 0.5g disodium
ethylene diamine tetra acetic acid (di-EDTA), 15.6g sodium dihydrogen
orthophosphate ((NaH2PC>4 ), pH3.8)) was mixed with 6pl of diluted sample in the
reaction cuvette. Fifty microlitres of start reagent (0.12% w/v hydrogen
peroxide) was added and the absorbance measured at 600nm every 25 seconds
until the reaction end point was reached using a MIRA discrete biochemistry
analyser (Roche Diagnostics, Welwyn Garden City, Herts). The assay was
calibrated using a bovine acute phase serum standardised against purified
haptoglobin.
79
3.2.1.2 Determination of Serum Amyloid A Concentration
To obtain sufficient quantities of serum amyloid A (SAA) to use as antigen in

monoclonal antibody production, serum was collected from diseased cattle
admitted to the University of Glasgow Veterinary School. Seventeen samples
with SAA levels greater than 50mg/l were identified by ELISA using a rabbit anti
human serum amyloid A antibody (Calbiochem-Novabiochem (UK) Limited,
Nottingham) (Table 3.1). High levels of SAA (>50mg/l) were measured in cattle
with a wide variety of clinical diseases (Table 3.1). The conditions that the cattle
were affected by included viral, bacterial and parasitic infections, tumours and
metabolic disease.
The sera were pooled and high density lipoprotein (HDL) with which SAA is
associated was isolated by ultracentrifugation thus removing the majority of serum
proteins. The SAA was further purified by electro-elution of the high density
lipoprotein fraction following sodium dodecyl sulphate polyacrylamide gel
electrophoresis (SDS-PAGE) while western blot analysis was used to confirm the
isolation and purification of the SAA.
3.2.2 Isolation and Purification of Serum Amyloid A
3.2.2.1 Separation of Serum from Blood
Blood samples from diseased cattle described in Table 3.1 were obtained by
venepuncture of the jugular vein. The sample was incubated at 37°C for 30
minutes to improve clot retraction and then at 4°C for a further 30 minutes. The
sample was then filtered to remove the clot and centrifuged at 1900g at room
temperature for 20 minutes. After centrifugation the serum fraction was carefully
removed, heat inactivated at 56°C for 30 minutes and stored at -20°C.
80
Case SAA
number Diagnosis (mg/1)
120942 Cellulitis 195.0
120898 Laryngeal abscess 52.6
120917 Mastitis and hepatitis 50.0
121121 Necrotising pneumonia 85.0
121194 Peritonitis and polyarthritis 92.3
121374 Intestinal carcinoma 68.1
121991 Meningitis and encephalitis 64.0
122058 Oral and interdigital ulceration 72.0
122105 Nephrosclerosis 77.5
122247 Hepatic lipidosis 76.1
122754 Lungworm 78.0
123302 Mucosal disease 82.0
123437 Mucosal disease 81.2
123589 Endocarditis 52.4
123669 Pyelonephritis 58.5
123745 Endocarditis and pulmonary thromboembolism 58.6
123811 Pulmonary thromboembolism/abomasal ulceration 72.3
Table 3.1: Serum amyloid A concentrations in diseased cattle.
81
3.2.2.2 Removal of Non-Lipoprotein Serum Proteins
To obtain serum amyloid A for use in the monoclonal antibody production,

several purification stages were carried out initially to isolate the high density
lipoprotein fraction and then the SAA from the other apolipoproteins.
To raise the non-protein density of the serum to 1.21g/ml, lOmls of serum was
mixed with 3.56g of potassium bromide and 6.5ml aliquots were placed in
thermoplastic ultracentrifuge tubes (Beckman Instruments UK Limited, High
Wycombe, Buckinghamshire). The tubes were capped, ensuring that air bubbles
were kept to a minimum, and centrifuged at 190,000g, at 4°C for 48 hours in a
Beckman L-70 Ultracentrifuge (fixed angle rotor type 50.4 Ti). Following
centrifugation, the top 1ml fraction of the samples containing the HDL was
carefully removed by pipette.
Potassium bromide was removed from the SAA fraction by dialysis. Dialysis
tubing D-9277 (Sigma) was soaked in distilled water for at least 60 minutes
before use. Fractions from the ultracentrifugation were dialysed against distilled
water at 4°C for 48 hours. Following dialysis, samples were removed from the
tubing and placed in sterile universals for lyophilisation. Samples were frozen
horizontally to achieve the largest surface area possible and then freeze-dried
overnight.
3.2.2.3 Determination of Total Protein Concentration of Serum
The total protein assay based on bicinchoninic acid reagent (Sigma) was used for
the determination of total serum protein and also for quantification of SAA after
purification. Bovine serum albumin (BSA, 2mg/ml in 0.15M NaCl with 0.05%
sodium azide) was used as a standard at a range of concentrations from 0-2mg/ml.
Dilutions were performed in distilled water. Ten microlitres of the standard and
82
the samples and 200pl of assay reagent (1 part copper (II) sulphate (pentahydrate
4% (w/v) solution ) mixed with 50 parts bicinchoninic acid solution) were
incubated in a 96-well mictotitre plate (Microtitre plates, Greiner) at 37°C for 30
minutes. Following incubation the absorbance was read at 562nm using a
MR5000 ELISA reader (Dynatech) and protein concentration calculated from the
BSA calibration curve.
3.2.2.4 Separation of Serum Amyloid A from High Density Lipoproteins
Further purification of SAA was required for monoclonal antibody production.

Following ultracentrifugation, dialysis and lyophilisation the semi-purified SAA
was still associated with HDL. Serum Amyloid A was separated from other HDL
apolipoproteins by SDS-PAGE followed by electro-elution in order to increase
the immunogenicity of the SAA, as fewer antigenic determinants of the SAA are
exposed in the native state (Godenir etal., 1985; Sipe etal., 1989).
SDS-PAGE gel and samples were prepared as described previously in the Large
Gel System. Electro-elution was performed using the Bio-Rad Model 422
Electro-eluter. Prior to electro-elution, membrane caps were placed in elution
buffer (25mM Tris-base, 192mM glycine and 0.1% SDS) at 60°C for at least 60
minutes. After heating, membrane caps were stored in elution buffer, which
contained 0.05% sodium azide as a preservative, at 4°C. After electrophoresis,
the gel was washed three times in deionised water and then stained for 10 minutes
using 0.1% coomassie brilliant blue in 25% methanol, 10% acetic acid and 1%
glycerol. The stained gel was washed in distilled/deionised water for two hours,
or until the protein bands became clearly visible. After washing the SAA band
was identified on the basis of molecular weight and a narrow strip of gel
incorporating the band was removed. The gel segment was then cut into 5mm x
5mm squares and stored in elution buffer at 4°C until electro-elution. The gel
squares were placed in the glass chambers within the electro-eluter module and
eluted at 8-10mA per chamber in protein elution buffer for approximately five
hours on ice, after which the sample was dislodged from the membrane caps by
83
rinsing with 200pl of fresh elution buffer and stored a 4°C until the concentration
stage. Alternatively, a narrow strip of stained gel was used as a template for
locating the SAA region from the unstained portion of a gel and samples were
electro-eluted as before.
3.2.2.5 Concentration of Serum Amyloid A
Concentration of the eluate from electro-elution was carried out using Centricon 3
Microconcentrators (Amicon Limited, Stonehouse, Gloucestershire) and stored at
-20°C. Eluate was tested using SDS-PAGE gel electrophoresis to determine
purity and the protein concentration was assessed using the BCA protein assay.
3.2.3 Identification of Serum Amyloid A
3.2.3.1 Enzyme Linked Immunosorbent Assay
To determine the SAA concentration in serum samples an ELISA was used

(Horadogoda et al, 1994). The standard used for this assay was a bovine serum
sample which had previously been quantified for SAA and haptoglobin levels.
The standard contained 200pg/ml of SAA and was diluted with fetal calf serum
(FCS) to give a range of SAA concentrations between 0-200pg/ml to establish a
calibration curve.
Standards, test samples and a control sample which was a ‘normal’ serum sample
containing negligible SAA were all diluted 1:400 in coating buffer (lOmM sodium
bicarbonate (NaHC0 3 )/sodium carbonate (Na2 CO3), pH 9.6). One hundred
microlitres was dispensed in duplicate into wells of a 96 well flat bottom
microtitre plate (Greiner Labortechnik Limited, Stonehouse, Gloucestershire).
The microtitre plate was incubated at room temperature (20-22°C) for eighteen
hours to allow the samples to bind to the plate. The fluid was then decanted and
the plates rinsed three times with phosphate buffered saline-Tween 20 (PBS-T,
84
20mM NaH2P 0 4 , 20 mM disodium hydrogen monophosphate (Na2 P 0 4 ), 154mM
sodium chloride (NaCl) and 0.05% (v/v) Tween-20 ), at pH 7.4.
To reduce non specific protein binding, 300(4.1 of 10% (w/v) dried skimmed milk
in PBS-T was added to each well and incubated at room temperature for sixty
minutes. Following incubation each well was washed three times with PBS-T.
Rabbit anti-human serum amyloid A (Calbiochem-Novabiochem (UK) Limited,

Nottingham) was prepared at a 1:10,000 dilution in PBS-T and lOOpl was added
to each well. The plate was then incubated at room temperature for ninety
minutes, and washed three times in PBS-T to ensure complete removal of any
excess primary antibody solution.
Horseradish peroxidase-conjugated donkey anti-rabbit IgG (Scottish Antibody

Production Unit, Law Hospital, Carluke) was prepared at a 1:2500 dilution in
PBS-T and lOOfil was added to each well, incubated at room temperature for
ninety minutes, and washed thoroughly three times in PBS-T.
Peroxidase substrate (0.01M acetate buffer, (pH5.5), 0.04% hydrogen peroxide,

and 0.1% TMB in dimethyl sulfoxide) was freshly prepared and 150pl was added
to each well and incubated at 37°C for 15 minutes. The substrate reaction was
terminated by the addition of 50|j.l of 2M sulphuric acid to each well.
The absorbance was measured at 450nm using a plate reader (Dynatech MR5000,
Dynatech Laboratories Limited, Billinghurst, West Sussex) and converted to SAA
concentration by the immuno assay software.
85
3.2.3.2 Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis.
Serum amyloid A was identified by molecular weight using sodium dodecyl

sulphate polyacrylamide gel electrophoresis (SDS-PAGE) stained with coomassie
brilliant blue. The identification of SAA was confirmed by immuno blot analysis
using antibodies specific for human SAA previously shown to cross react with
bovine SAA (Horadogoda, 1994).
Mini-Gel System. Standards, controls and test samples were diluted 1:20 in
distilled/deionised water. The wide range molecular weight marker (M.W. 6,500-
205,000, Sigma-Aldrich Company Limited, Poole, Dorset) and samples were
mixed 1:1 with treatment buffer (0.125M Tris-Cl, (pH 6.8), 4% sodium dodecyl
sulphate (SDS), 20% glycerol and 10% 2-mercaptoethanol) and boiled at 100°C
for 2-3 minutes to denature the protein structure by reduction of disulphide
bonds. Samples were then centrifuged at 13800g using a Eppendorf 5415
centrifuge (Eppendorf limited, Hamburg) for 30 seconds.
SDS-PAGE resolving gel solution (0.375M Tris-HCl, (pH8.8) 0.1% SDS; 0.1%
(w/v) ammonium persulphate (APS); 0.07% N, N, N ’, N ’-tetremethyl-
ethylenediamine (TEMED) and 15% bis acrylamide) was prepared and added to
the gel apparatus (8.0cm x 7.3cm x 0.75cm, Biorad Minigel System, Biorad
Laboratories Limited, Hemel Hempstead) adding a barrier layer of 70% propanol
to aid the formation of the gel matrix. When the resolving gel had solidified at
room temperature, the gel was washed with distilled water to remove the
propanol. The stacking gel (0.125M Tris-HCl, (pH6.8) 0.1% SDS; 0.05% (w/v)
APS, 0.05% TEMED, 4% bis acrylamide) was added, inserting the combs
carefully. The combs were removed, gel apparatus was placed in the gel tank
(Biorad - Mini Protean II cell) and the central well filled with tank buffer (0.025M
Tris, (pH8.3), 0.192M glycine and 0.1% SDS). Tank buffer was also added into
the buffer tank. Fifteen microlitres of molecular weight marker and 20 jj1 of each
sample was loaded into each gel and electrophoresed at 200V for 60 minutes.
86
After electrophoresis, one gel was removed and placed in coomassie stain
(0.125% coomassie blue R-250, 50% (v/v) methanol, 10% (v/v) acetic acid) for
5-10 minutes, followed by destaining in destain (5% (v/v) methanol, 7% (v/v)
acetic acid) for 20 minutes, while a duplicate gel was processed by
immunoblotting.
Large Gel System. The method used was similar to the mini-gel system except
that the gel dimensions used were 16cm x 20cm x 1.5mm and larger sample
volumes (60pl per well) were loaded. The gel was electrophoresed for three
hours at 200 Volts.
3.2.3.3 Immuno Blotting
The duplicate SDS-PAGE gel (unstained) was used for the Immuno Blotting
technique. Nitrocellulose membrane (Transblot Transfer medium, 0.45pm,
Biorad) (1 piece) and filter paper (1 piece) were prepared before use by soaking in
transfer buffer (25mM tris, 192mM glycine and 20% (v/v) methanol).
The SDS-PAGE gel was placed between the filter paper and nitrocellulose
membrane in the transfer plates. The plates were covered with transfer buffer and
electrophoresed at 100V for 30 minutes. After transfer, the nitrocellulose
membrane was blocked overnight with 5% (w/v) dried skimmed milk solution in
Tris buffered saline-Tween 20 (TBS-T, lOmM Tris-Cl (pH8.2), 150mMNaCl and
0.5% Tween 20) on a rotatest shaker (Luckham R100, Luckham Limited, Sussex)
at 4°C. After blocking, the nitrocellulose membrane was rinsed three times in
approximately 50mls of TBS-T. Fifty millilitres of antibody solution (rabbit anti
human SAA, Calbiochem) was added at 1:1000 in TBS-T, and incubated on a
rotatest shaker for 60 minutes at room temperature. The antibody solution was
then removed and the nitrocellulose membrane was rinsed three times with TBS-
T. To ensure complete removal of excess antibody solution a further two, 10
minute, washes in 50mls of TBS-T were carried out. Fifty millilitres of secondary
87
antibody (donkey-anti-rabbit IgG labelled with horseradish peroxidase) was added
at 1:1000 in TBS-T and incubated for 60 minutes on the shaker. The
nitrocellulose membrane was then rinsed and washed in TBS-T as above. Amino
ethyl carbazole (AEC) substrate (20mg AEC dissolved in 2.5ml dimethyl
formamide and added to 0.05M acetate buffer (pH5) to give a final volume of
50ml, plus lOOpl of 30% hydrogen peroxide solution) was freshly prepared and
poured onto the blot and incubated for 5-15 minutes. The blot was then washed
in distilled water to terminate the reaction and dried.
3.2.3.4 Immuno Dot-Blot
The immuno-dot blot was used as a rapid, semi-quantitative method for

identifying samples with elevated SAA. Ten microlitre aliquots of the samples
and the standards were added in duplicate to nitrocellulose membrane and allowed
to air dry at room temperature. The nitrocellulose membrane was blocked with
5% (w/v) dried skimmed milk in TBS-T for two hours on a rotatest shaker at
room temperature. Following incubation the method was as described previously
for the immunoblotting technique (after the blocking procedure stage).
3.2.4 Identification and Quantification of Acute Phase Proteins in

Milk Samples from Mastitic Cows
3.2.4.1 Determination of Haptoglobin Concentration
To determine if acute phase proteins could be detected in milk, the concentration

of haptoglobin was determined using a modification of the semi-automated
haptoglobin assay. Stock solutions were used at concentrations described as for
detection of haptoglobin from bovine sera unless otherwise stated. The
haptoglobin standard (bovine acute phase serum of known haptoglobin
concentration) was diluted in normal serum (negligible haptoglobin) to give a
range of concentrations from 0.3mg/ml to 3.0mg/ml. Twenty microlitres of the
88
standards, controls and samples were then placed in test tubes, IOOjj,! of standard
haemoglobin solution was added and the samples were mixed thoroughly and
incubated at room temperature for 10 minutes. Five millilitres of 0.9% saline was
added to the tubes, mixed, and then 20jj.l of the sample was pipetted in duplicate
into a 96-well microtitre plate (Greiner). Two hundred microlitres of chromogen
solution was added and the plate incubated at 37°C for 60 minutes. Fifty
microlitres of substrate was added and the samples were then incubated at room
temperature until adequate colour development had occurred in the standard
samples. The reaction was terminated by adding 50|il of 20% sulphuric acid and
the absorbance was read at 450nm using an MR5000 plate reader (Dynatech).
The haptoglobin concentrations were calculated by comparison to the standard
haptoglobin solution.
3.2.4.2 Determination of Serum Amyloid A Concentration
SAA concentrations in milk samples were determined by ELISA as described

previously for SAA in bovine sera. The concentrations of samples, primary
antibody and secondary antibody were varied to determine the optimum assay
conditions. Due to the original antibody being unavailable due to an alteration in
the manufacturing procedures, other possible antibodies had to be assessed to
attempt to and find a suitable replacement for detecting bovine SAA. The
alternative antibodies were assessed initially with a serum standard with a known
SAA concentration by ELISA, SDS-PAGE and immunoblotting, using the
method described for measurement of SAA in sera. Primary antibody
concentrations varying from 1:1000 to 1:20,000 were incubated with samples
diluted in coating buffer from 1:100 to 1:1,200 to try to optimise the assay. The
antibodies assessed by ELISA, SDS-PAGE and immunoblotting were: rabbit anti
human serum amyloid A component, non-precipitating, (Calbiochem) and a rat
monoclonal anti human SAA, (2° antibody-rat anti IgG HRP labelled, (Biosource
International, Serotec, Kidlington, Oxford). A rabbit anti-bovine AA and a rabbit
anti-bovine SAA kindly provided by Professor Gruys, Utrecht University, The
Netherlands were analysed by SDS-PAGE and immunoblotting.
89
3.2.5 Development of a Monoclonal Antibody to Serum Amyloid A
The techniques attempting to develop a monoclonal antibody to purified SAA

were carried out by Rhone Poulenc Diagnostics Limited, West of Scotland
Science Park, Maryhill Road, Glasgow. Collaboration with colleagues from
Rhone Poulenc took place throughout the purification stages and the
immunisation procedure for antibody production.
3.2.5.1 Immunisation of Mice with Bovine Serum Amyloid A
The initial immunisation procedure used purified SAA at a concentration of

180(ig/ml in elution buffer (24mM tris, 0.19M glycine and 0.02% SDS). The
SAA was obtained by the electro-elution method described previously where the
SDS-PAGE gel was partially stained with coomassie to identify the SAA rich
portion. This resulted in the coomassie stain binding to the SAA which remained
even following elution.
A group of 3 BALB/c mice were given a subcutaneous injection of lOOpl of

immunogen which contained 50}ig of antigen in an emulsion prepared with
complete Freund’s adjuvant. Twenty one days later, lOOpl of immunogen
containing 20fig of antigen in an emulsion of incomplete Freund’s adjuvant was
injected subcutaneously and repeated 14 days later. Approximately 7 days after
the third subcutaneous immunisation a test bleed was carried out to determine if
an antibody response had occurred. A sample of blood was removed from the tail
vein and allowed to clot, serum was removed after centrifugation and the serum
antibody titre tested by ELISA with SAA bound to the wells of the microtitre
plates.
In monoclonal antibody production if a good antibody response occurs the

animals are rested and then boosted by a intravenous injection of antigen in saline,
with the removal of the spleen and blood occurring 4 days later, followed by
90
fusion of spleen cells. However if a poor antibody response occurs then re-
immunisation with antigen is generally performed.
Due to the poor antibody response of the mice after the initial immunisation with
SAA (coomassie bound) a second immunisation with purified SAA (184pg/ml in
elution buffer) which had been isolated using the alternative electro-elution
method, and therefore, free of the coomassie stain, was carried out.
3.2.5.2 Immunisation of Mice with Bovine Serum Amyloid A-Apoferritin

Complex
SAA is a relatively small molecular weight protein (approximately 12.5kDa). A

good immune response to the antigen may not occur due to poor antigenicity of a
small protein. To improve the immune response of small immunogens a carrier
protein may be attached which will increase the size of the immunogen and may
provide the binding sites essential for an immune reaction. An additional
immunisation technique used SAA bound to apoferritin, a carrier protein isolated
from horse spleen, to try to improve the immune response of the animals to the
SAA by increasing the molecular mass of the immunogen.
Purified SAA was desalted and concentrated using centricon 3

microconcentrators (Amicon Limited) with 50mM triethanolamine, 0.15M sodium
chloride, 2mM EDTA, (pH 8.0) to a final volume of 0.3ml. Freshly prepared 2
iminothiolane, a protein modification reagent, was mixed with bovine SAA under
nitrogen at room temperature for 45 minutes and desalted through a precalibrated
PD10 column of G25 sephadex (Pharmacia Limited, Milton Keynes) with 0.083M
sodium phosphate, 0.9M sodium chloride, 0.1M EDTA, (pH7.2) and the eluted
peak of labelled SAA collected.
Two milligrams of the carrier protein apoferritin was mixed with 300pi of
PBS/EDTA. Four-(N maleimidomethyl) cyclohexane carboxylic acid N-hydroxy
succinimide ester (SMCC) was mixed with dimethylsulfoxide (DMSO) and
91
combined with the apoferritin solution at room temperature for 30 minutes and
desalted again through a PD10 column and the purified peak of apoferritin
collected. The desalted fractions of modified SAA and apoferritin were mixed
and allowed to react overnight at 4°C. All traces of EDTA were removed from
the protein conjugate by centrifugation in centricon 3 microconcentrators and
buffer exchange with PBS until a final volume of 1ml was obtained.
The immunisation procedure was similar to the initial protocol described above,
with the exception that 4 BALB/c mice were immunised with bovine SAA-
apoferritin complex. The first and second immunisations involved the
subcutaneous injection of lOOpl of immunogen containing 100|ig of apoferritin
and 20fig SAA prepared as an emulsion in Freund’s adjuvant (1st immunisation -
complete adjuvant, 2nd - incomplete adjuvant). A third subcutaneous injection
containing 50|ig of apoferritin and lOfig of SAA in Freund’s incomplete adjuvant
was carried out. A test bleed to determine the antibody titre was taken before the
animals were boosted, with a 200pl intravenous injection containing 25jig of
apoferritin and 5fig of SAA in saline.
3.2.5.3 Measurement of Antisera Response to Immunisation of Mice with

Bovine Serum Amyloid A
An ELISA assay was developed to measure the antisera response to immunisation

of mice with bovine serum amyloid A. Microtitre plates were coated with purified
bovine SAA and the serum response measured from the immunised mice using
non-immunised mouse serum as a control. The bound antibody was detected
using an anti-mouse IgG-horseradish peroxidase labelled antibody (Sigma) with
TMB as a substrate.
The immuno dot blot technique was used as a more sensitive method of
determining the antibody response of the mice. The streptavidin/biotin system
was used to determine if any significant response had occurred as this system is
more sensitive than other detection procedures and can be used accurately for the
92
detection and localisation of antigens, glycoconjugates and nucleic acids. Dot
blot analysis was carried out using 2\x\ sample volumes with serial dilutions of the
purified standard SAA, high and low serum SAA standards and 1:250 dilutions of
the mouse control serum and test samples. Anti-mouse IgG biotin labelled
antibody (Sigma) was used followed by streptavidin labelled HRP. Bound
peroxidase was detected after washing with the substrate 4 chloro, 1 naphthol.
3.3 Results
3.3.1 Identification and Quantification of Acute Phase Proteins

from Sera of Diseased Cattle
A bovine serum sample of known SAA concentration was used as a standard

allowing construction of a calibration curve and determination of the SAA
concentration in the unknown serum samples using a polyclonal rabbit anti-human
SAA antibody. Cattle affected by conditions which the farm animal clinicians at
Glasgow Veterinary School considered to involve a significant inflammatory
response failed to consistently show elevated levels of SAA in ELISA using the
polyclonal rabbit anti-human SAA antibody. It was, therefore, not possible to
select animals with high SAA on clinical criteria. Previous work by Horadogoda
et al. (1994) had shown that SAA levels were related to haptoglobin levels in
cattle, but that the SAA was raised in a number of samples where haptoglobin was
normal. However, haptoglobin levels in these selected cases varied from normal
(less than 0.1 mg/ml) to increased (3.06mg/ml) indicating that haptoglobin levels
were not consistently elevated in all animals with high SAA. Blood samples from
all cattle cases were subsequently screened for SAA and cases were then selected
on the basis of having a SAA concentration of > 50mg/l and these included a
variety of infectious and inflammatory conditions (Table 3.2).
93
3.3.2 Determination of Serum Amyloid A Concentration
A calibration curve for bovine SAA measured by ELISA was constructed by

dilution of an acute phase serum standard with a SAA concentration of 192pg/ml
in FCS to give a range of concentrations from approximately 0-192pg/ml (Table
3.3; Figure 3.1).
3.3.2.1 Isolation and Purification of Serum Amyloid A
In order to obtain SAA for use in monoclonal antibody production, the SAA had
to be purified, which involved a number of isolation stages and confirmation of
the presence of the SAA after each purification step.
The initial purification of SAA from serum involved ultracentrifugation of the

samples which separated the serum into density determined fractions containing
very low density proteins, low density proteins, medium density proteins and high
density proteins which were designated fractions 1-4 respectively. The BCA total
protein assay gave typical results as shown in table 3.4 and in figure 3.2 and was
used for determination of total protein content of fractions 1-4 (Table 3.5).
3.3.2.2 Identification of Serum Amyloid A
To assess the presence of SAA in serum, SDS-PAGE was used to identify the
SAA, which is a 12.5 kilodalton protein (Betts et al., 1991) and can be identified
by comparison with molecular weight (MW) standards which ranged from M.W.
6,500 to 205,000. The presence of SAA was also confirmed by comparison with
serum samples known to contain negligible, or high levels of SAA, which were
used as negative and positive controls respectively.
Following identification of SAA by molecular weight, the presence of SAA was

confirmed by immunoblotting using the polyclonal rabbit anti-human SAA
94
Case SAA Hp
number Diagnosis (mg/1) m
120942 Cellulitis 195.0 1.4
120898 Laryngeal abscess 52.6 <0.06
120917 Mastitis and hepatitis 50.0 <0.06
121121 Necrotising pneumonia 85.0 0.8
121194 Peritonitis and polyarthritis 92.3 <0.06
121374 Intestinal carcinoma 68.1 0.06

121991 Meningitis and encephalitis 64.0 0.24
122058 Oral and interdigital ulceration 72.0 <0.06
122105 Nephrosclerosis 77.5 0.36
122247 Hepatic lipidosis 76.1 <0.06
122754 Lungworm 78.0 2.34
123302 Mucosal disease 82.0 1.68
123437 Mucosal disease 81.2 2.34
123589 Endocarditis 52.4 1.92
123669 Pyelonephritis 58.5 3.06
123745 Endocarditis and pulmonary thromboembolism 58.6 <0.06
123811 Pulmonary thromboembolism/abomasal ulceration 72.3 1.62
Table 3.2: Serum haptoglobin concentrations in diseased cattle with SAA levels
>50mg/l referred to Glasgow University Veterinary School. SAA = serum
amyloid A. Hp = haptoglobin, haptoglobin reference range < 0.3mg/ml for
normal animals.
95
SAA Concentration (^ig/ml) Mean Absorbance (450nm)
192 0.951
96 0.691
48 0.536
19.2 0.402
9.6 0.230
0 0.047
Table 3.3: Standard SAA (concentration 192|ig/ml) used for the construction of a
calibration curve for bovine SAA measured by ELISA. The average absorbances
of 10 curves at 450nm are shown. SAA = serum amyloid A.
96
97
(oiuosfr) eoueqjosqe
o
LO
CM
Figure 3.1: Calibration curve for Bovine SAA

BSA Concentration Mean Absorbance (562nm)
(mg/ml)
2.0m g/ml 0.997

1.5mg/ml 0.830
1.0 mg/ml 0.606
0.5mg/ml 0.352
0.0 0.050
Table 3.4: Standard BSA (concentration 2.0mg/ml) used for the construction of a
calibration curve for total protein measured by BCA total protein assay. The
average absorbances of 10 curves at 562nm are shown. BCA = bicinchoninic acid
98
Total Protein Calibration Curve
04
o
o
99
o
o
04
(ujuggg) eoueqjosqe
O
O
lO
ID
CM
o
o
o
c
c
c
<
o
+ -•
00
E
a)
CO
O)
aS
Figure 3.2: Calibration curve for BCA total protein a ssa y

Sample Dilution Average Total Protein
Absorbance Concentration
(562nm) (g/1)
Fraction 1 1:10 0.354 5.08
Fraction 2 UNDIL 0.690 1.17
Fraction 3 1:100 0.445 67.10
Fraction 4 1:1000 0.299 412.00
Table 3 .5: The average absorbances for each fraction following ultracentrifugation
are shown. Total protein concentration calculated from BCA total protein assay
using BSA calibration data. BSA = bovine serum albumin. UNDIL = undiluted
sample. BCA = bicinchoninic acid
100
antibody. Positive and negative control sera were also immunoblotted to confirm
the specificity of the rabbit anti-human SAA for bovine SAA.
The SAA with a molecular weight of approximately 12,500 Daltons was identified
in fraction 1 (high density lipoprotein fraction (HDL)) after SDS-PAGE (Figure
3.3) and immunoblotting (Figure 3.4). SAA was also identified in fraction 2
(medium density proteins) at a lower concentration. Fractions 3 and 4 (low and
very low density proteins respectively) contained no SAA.
The SAA fraction was further purified by dialysis to remove any remaining K-Br
and then freeze dried. The presence of SAA was again confirmed by SDS-PAGE
(Figure 3.5) and immunoblotting (immunoblots not shown).
Electro-elution of the SAA containing fraction which had been semi-purified by

ultracentrifugation, dialysis and lyophilisation was used to further purify the SAA.
The electro-elution was carried out on three separate occasions to further purify
three batches of semi-purified SAA. Following electro-elution the SAA was
concentrated to approximately 200pg/ml, determined using the BCA total protein
assay (Table 3.6). The purified SAA was then checked for purity by SDS-PAGE
and immunoblotting where the band isolated showed positive staining as in Figure
3.4 (results not shown).
3.3.3 Development of a Monoclonal Antibody to Bovine Serum

Amyloid A
3.3.3.1 Immunisation of Mice with Bovine Serum Amyloid A
The sera from mice immunised with SAA were analysed using an immuno dot blot
and the streptavidin/biotin system. No colour development was detected in serum
samples from mice immunised with the purified bovine SAA, at any of the
dilutions of sera assessed.
101
MW
kd
2 0 5 .0
4 0 .0
1 4 .0
6 .5
3 4 5
Figure 3.3: SDS-polyacrylamide gel (15%) electrophoresis of acute phase bovine

serum fractions following ultracentrifugation. Lane 1 shows fraction 1 (HDL
fraction containing SAA). Lane 2 shows fractions 2 (medium density proteins)
which contains SAA at a lower concentration than fraction 1. Lanes 3 and 4
show fractions 3 and 4 respectively which contain no SAA. Lane 5 shows the
standard serum which contained SAA at a concentration of 192pg/ml.
102
MW
kd
1 4 .0
6.5
Figure 3.4: Immunoblot after SDS-PAGE of acute phase bovine serum fractions
following ultracentrifugation. Polyclonal rabbit anti-human SAA was used as the
primary antibody. Lane 1 shows the standard serum which contained SAA
(concentraion 192p.g/ml). Lane 2 shows fraction 2 (medium density protein
fraction) which contains SAA. Lane 3 shows fraction 1 (HDL fraction) with the
SAA region clearly identified by the immunoblot.
103
1
MW
kd
2 0 5 .0
400
14.0
6 .5
Figure 3.5: Coomassie stained SDS-polyacrylamide gel (15%) electrophoresis of

the HDL portion of acute phase serum following ultracentrifugation, dialysis and
lyophilisation at a range of dilutions. Lanes 1-4, shows the fraction 1 with SAA
identified by comparison with the MW marker at dilutions of fraction 1 of 1:16,
1:8, 1:4 and 1:2 respectively. Lane 5 shows the standard SAA (concentration
192|ig/ml).
104
Electro-elution Average Absorbance Total Protein
(562nm) (Hg/ml)
1 0.228 227
2 0.297 229
3 0.231 184
Table 3.6: Total protein concentration of electro-eluates 1, 2 and 3 using the

BCA total protein assay.
105
3.3.3.2 Immunisation with Bovine Serum Amyloid A-Apoferritin Complex
The sera from mice immunised with serum SAA-apoferritin complex was analysed
using ELISA. No colour development was detected in serum samples from mice
immunised with the purified bovine SAA-apoferritin complex, at any of the
dilutions of sera assessed.
These two approaches to produce a monoclonal antibody to bovine SAA were

unsuccessful and it was not possible to attempt further immunisation regimes
within the duration of the project.
3.3.4 Identification and Quantification of Acute Phase Proteins in

Milk Samples From Mastitic Cows
Milk samples were taken from the four cows challenged by intramammary
infusion with S. aureus (cows 125, 186, 520 and 703), but only from the hind
quarters (LH and RH) on four separate dates, i.e. days 0 (pre-challenge), 4, 8 and
17 post-challenge.
3.3.4.1 Quantification of Total Protein and Haptoglobin
The skimmed milk fraction was analysed to determine the total protein
concentration using the total protein assay described for serum. The same
samples used for total protein measurements were also used to assess haptoglobin
levels (Tables 3.7 - 3.10). Normal serum haptoglobin levels are <0.10g/l and a
serum haptoglobin level of >0.40g/l is indicative of infection (Makimura and
Suzuki, 1982; Skinner et a l , 1991). The haptoglobin assay detection limit was
between 0.10g/l and 0.14g/l.
All quarters sampled in three of the four cows on all sample dates, had either very
low concentrations of haptoglobin or levels of haptoglobin which were below the
limit of detection of the assay (tables 3.7-3.9). A haptoglobin concentration of
106
0.50g/l was measured on day 8 post-challenge in the LH quarter of cow 703
which was non-infected and which had been challenged with strain B. The
concentration of haptoglobin in this quarter was not elevated on day 4 and day 17
post-challenge (table 3.10).
3.3.4.2 Detection of Serum Amyloid A
Alternative antibodies were assessed for their ability to detect bovine SAA due to
the unavailability of the original rabbit anti-human SAA from Calbiochem. The
polyclonal rabbit anti-human serum amyloid A component (non-precipitating)
(Calbiochem) and monoclonal rat anti-human SAA (Biosource International)
antibodies were assessed by ELISA, SDS-PAGE and immunoblotting using serum
standards with known concentrations of SAA. Both antibodies gave a
disappointing result with the ELISA showing no significant increase in absorbance
in the positive control serum compared to the negative control serum. High
background also occurred and this was reduced by changing the blocking agent to
2% BSA in PBS-T, and diluting the primary antibody in 0.1% BSA in PBS-T.
Following SDS-PAGE and immunoblotting, the monoclonal rat anti-human SAA
gave a weak positive result for all the proteins in the serum sample and was not
specific for bovine SAA. The polyclonal rabbit anti-human SAA component,
(non-precipitating) antibody was also shown to be non-specific for bovine SAA.
The monoclonal rabbit anti-bovine AA and the monoclonal rabbit anti-bovine
SAA antibody (kindly donated by Professor Gruys, Utrecht, Netherlands) both
gave very poor responses to bovine SAA on immunoblotting, however a weak but
specific response to bovine SAA on immunoblotting was just detectable using the
rabbit anti-bovine SAA.
Due to the lack of a reliable antibody for detection of bovine SAA, SDS-PAGE
was performed to determine whether SAA was present in milk. As SAA has a
characteristic MW of approximately 12,500 and can be observed in SDS-PAGE
gels of acute phase serum, identification of a protein of identical size to SAA not
present in normal milk, but detectable in milk after infection or inflammation could
suggest the presence of SAA in milk. Initially, cow 703 was studied as she had
107
Cow Sample LH Q uarter RH Q uarter
Number day Infected Strain A Control
Challenged Strain A
Haptoglobin Total protein Haptoglobin Total protein
(mg/ml) (mg/ml) (mg/ml) (mg/ml)
125 00 < LD 8.40 0.16 12.20

125 04 < LD 12.40 < LD 11.50
125 08 0.20 9.10 0.17 9.80
125 17 < LD 7.30 0.15 9.70
Table 3.7: Haptoglobin and total protein concentrations identified in mammary

secretions from the hind quarters of cow 125 days 0 (pre-challenge), 4, 8 and 17
post-challenge. LD indicates limit of assay detection which is <0.10g/l. LH = left
hind. RH = right hind.
108
Number Day Infected Strain A Control
Challenged Strain B
(rag/ml) (mg/ml) (mg/ml) (mg/ml)
186 00 0.12 5.00 0.15 13.50

186 04 0.15 11.90 0.18 13.90
186 08 0.18 11.70 0.15 6.2
186 17 0.19 17.30 0.21 16.00
Table 3.8: Haptoglobin and total protein concentrations identified in mammary

secretions from the hind quarters of cow 186, days 0 (pre-challenge), 4, 8 and 17
post-challenge. LH = left hind. RH = right hind.
109
Number day Control Infected Strain B
Challenged Strain A
520 0 0.14 14.00 0.18 12.70

520 4 0.14 11.70 0.19 14.20
520 8 0.15 14.50 NS NS
520 17 0.13 13.70 0.19 20.60
Table 3.9: Haptoglobin and total protein concentrations in mammary secretions

from the hind quarters of cow 520 days 0 (pre-challenge), 4, 8 and 17 post
challenge. LH = left hind. RH = right hind. NS = no sample.
110
Number day Challenged Strain B Control

703 0 NS NS 0.16 20.05

703 4 0.22 15.50 0.22 14.25
703 8 0.50 26.80 NS NS
703 17 0.17 14.2 0.13 11.5
Table 3.10: Haptoglobin and total protein concentrations in mammary secretions

from the hind quarters of cow 703 days 0 (pre-challenge), 4, 8 and 17 post
challenge. LH = left hind. RH = right hind. NS = no sample.
I ll
shown an increase in haptoglobin levels in the LH quarter on day 8 post
challenge. Samples from both the LH and RH quarters were analysed by SDS-
PAGE from the four sampling dates i.e., days 0 (pre-challenge), 4, 8 and 17 post
challenge.
A protein of similar molecular weight to SAA was detected in the LH quarter of

cow 703 on day 8 post-challenge and this protein was not present in any other
sample days (Figure 3.6; lane 5). The ‘suspected’ SAA was identified by
molecular weight and compared to the semi-purified SAA, confirming the
presence of a protein of similar MW to SAA in milk samples. This approach was
thought to be as accurate as possible in the absence of a specific antibody for use
in confirming SAA by immunoblotting. To confirm that this additional protein
band was probably SAA a negative control milk sample was mixed with a range
of concentrations of SAA purified from serum. An additional band was observed
at SAA concentrations of 120mg/l and 60mg/l which was at a similar position in
the SDS-PAGE milk gel as the additional protein previously identified (figure
3.7). Figure 3.8 shows a negative control milk sample from the RH quarter on
day 3 post-challenge and a milk sample from the LH quarter of cow 703 on day 8
post-challenge which contains a protein of similar molecular weight to SAA. The
purified SAA and the milk sample which was mixed with the serum SAA all show
a protein with similar molecular weight to SAA.
112
MW
kd
2050
4 0 .0
1 4.0
6 .5
Figure 3.6: Coomassie stained SDS-polyacrylamide gel (15%) electrophoresis of

milk samples from cow 703 on days 0, 4, 8 and 17 post-challenge from the LH
and RH quarters. Semi-purified SAA was used as a control in the identification of
similar molecular weight proteins. Lane 1 shows MW marker. Lane 2 shows RH
quarter day 3 post-challenge. Lanes 3 and 4 show LH and RH quarters
respectively from day 4 post-challenge. Lane 5 shows LH sample from day 8
post-challenge. Lane 6 and 7 show LH and RH samples from day 17 post
challenge. Lane 8 shows semi-purified SAA. T indicates the ‘suspected SAA’.
113
MW
kd
2050
4 0 .0
1 4 .0
6 .5
Figure 3.7: Coomassie stained SDS-polyacrylamide gel (15%) electrophoresis of a

control milk sample ‘spiked’ with a range of concentrations of semi-purified SAA.
Lane 1 shows MW marker. Lane 2 shows semi-purified SAA. Milk samples
‘spiked’ with SAA at concentrations of 120mg/l (lane 3) and 60mg/l (lane 4).
Milk samples ‘spiked’ with SAA at concentrations of 30 (lane 5), 15 (lane 6) and
Omg/1 (lane 7) respectively. T indicates the ‘suspected SAA’
114
MW
kd
2 0 5 .0
4 0 .0
1 4 .0
*
6 .5
Figure 3.8: Coomassie stained SDS-polyacrylamide (15%) electrophoresis. Lane

1 shows MW marker. Lanes 2 shows a negative control milk sample (RH, day 3
post-challenge). Lane 3 shows semi-purified SAA. Lane 4 shows a suspected
SAA milk sample (LH, cow 703, day 8 post-challenge). Lane 5 shows a ‘spiked’
milk sample with an SAA concentration of 120mg/l. T indicates the ‘suspected
SAA’.
115
I
3.4 Discussion
Identification o f the presence o f pathological lesions resulting from infectious or

inflammatory disease in cattle has been shown to be possible by monitoring APP
such as SAA and haptoglobin (Alsemgeest et a l, 1994; Gruys et a l, 1993).
Measurement o f APP, which remain elevated for hours or days, to quantify
disease activity is more useful than measuring cytokines which are rapidly cleared
from the circulation (Hoi et al., 1987). Assays for APP, particularly SAA, may
potentially be used to provide aid in prognosis and early diagnosis o f conditions
such as bovine mastitis (Conner and Eckersall, 1986) and pneumonic
pasteurellosis (Horadogoda et a l , 1993; Horadogoda et al., 1994). Serum
amyloid A and haptoglobin were found to be elevated in most animals which had
a history of disease at point of slaughter (Gruys et a l, 1993) and may, therefore,
potentially be useful in future monitoring of the health status of slaughter cattle
and in meat inspection (Saini and Webert, 1991).
Bovine milk contains a number of plasma proteins including albumin, transferrin,

ceruloplasmin and various immunoglobulins but prior to this investigation it was
not known if any SAA or haptoglobin could be detected in mammary secretions.
However, it is known that haptoglobin acts as an inflammatory marker in serum in
E. coli mastitis which is an acute, usually self-limiting mastitis (Salonen et al,
1996). A 52-fold increase in serum haptoglobin was measured after the first
intramammary challenge with E. coli and the peak concentration occurred 48-120
hours after the challenge (Salonen et a l, 1996). Thus, an APR was induced by E.
coli mastitis and it was o f interest to establish if an APR could be identified in
milk following S. aureus infection in the present study.
SAA binds bacterial products including lipopolysaccharides (Tobias et a l, 1982),

harmful molecules and debris produced after tissue damage (Baumann and
Gauldie, 1994). Lipopolysaccharides are produced by Gram negative mastitis
organisms especially the coliforms (Lipman et a l, 1995). Following intradermal
or intravenous administration o f endotoxin the concentration of SAA increased
116
100-fold (Boosman et al, 1990). Hirvonen et a l (1996) studied the APR in
heifers with experimentally induced mastitis, by measurement of serum
haptoglobin, acid soluble glycoproteins, plasma fibrinogen and alpha-1 -proteinase
inhibitor. They concluded that haptoglobin and acid soluble glycoproteins were
the most effective in indicating the severity of infection and predicting the final
outcome of the disease, whereas alpha-1 -proteinase inhibitor activity was of low
diagnostic value in this study. The measurement of alpha-1- antitrypsin and
bovine serum albumin in milk as an indicator of mastitis was carried out by
Sandholm et a l (1984), who concluded that alpha-1-antitrypsin was an effective
indicator of mastitis. Recent studies at Glasgow Veterinary School indicate that
the SAA is a more sensitive marker for inflammation than haptoglobin (Eckersall,
personal communication). Therefore, a major aim of this study was to identify
and quantify SAA in mastitic milk and a monoclonal antibody raised specifically
against bovine SAA was desirable for the development of a standard detection
assay, as alternative antisera had proved to be unreliable, and ultimately to be
unavailable. In order to achieve this, bovine SAA had to be accurately identified
and purified before being used as the antigen in monoclonal antibody production.
Identification of bovine clinical cases with elevated SAA concentrations proved to
be difficult, as cattle which were considered to be affected by clinically-
recognisable inflammatory conditions did not have consistently high levels of
SAA. The previously reported correlation between elevated haptoglobin and
SAA concentrations reported by Horodagoda et a l (1994) was not consistently
observed in the cases studied during this project. However, bovine SAA was
identified from a proportion of randomly selected clinical cases including those
with infectious and inflammatory conditions, using, a commercial polyclonal rabbit
anti-human antibody. The specificity of reaction with this antibody was confirmed
originally by staining of a low molecular weight band of approximately 12,500
Daltons, which was consistent with the SAA band identified in previous studies.
Sera with high concentrations of SAA were pooled, and SAA was then purified
by ultra-centrifugation, dialysis, lyophilisation, electrophoresis and electro-elution.
The purification of SAA by these methods resulted in isolation of purified bovine
SAA with a concentration of approximately 200|ig/ml. Development of this
117
method during the project provided an alternative to the hydrophobic interaction
chromatography and repeated gel filtration previously performed by Horadagoda
et a l (1993) in the purification and quantitative measurement of bovine SAA
Thus, the newly developed method did not require the high ionic strength
guanidium chloride, used previously to desalt the SAA and to keep it solubilised.
The gel filtration step had previously led to high loss of protein and was avoided
in the current method. Hydrophobic interaction chromatography was also used as
a purification method for human SAA by Raynes and McAdam (1988) with
affinity chromatography used in the preparation of hamster SAA (Niewold and
Tooten, 1991). Rossevatn et a l (1992) purified SAA by isolating the HDL
fraction by ultracentrifugation followed by de-lipidation and gel filtration.
Purification of proteins from polyacrylamide gels (Harlow and Lane, 1988) is a
standard method in parasitic antigen separation (McKeand, 1992) and purification
of proteins by electro-elution for subsequent use in antibody production has been
successfully carried out by many authors including Kennedy et a l (1986) and
Tomlinson et a l (1989). Antigens purified from polyacrylamide gels often induce
good antibody responses even though some denaturation of the antigen occurs.
Indeed, many molecules become more immunogenic following denaturation, as
injecting denatured antigens is more likely to produce an antibody response
against epitopes that are not found on the native antigen (Harlow and Lane,
1988).
In this study, the production of a monoclonal antibody to bovine SAA was

unsuccessful. No response was observed in the sera from mice immunised with
bovine SAA and analysed by immunodot-blot. Unfortunately, it was not possible
to attempt further immunisation required for monoclonal antibody production
during this short project. Further attempts at antibody production should examine
alternative strategies, including the effect of varying dose of immunising antigen,
the effect of the isolation procedure on the native state of the protein which may
have rendered it non-immunogenic (Zola, 1995), and the strain of mouse used for
monoclonal antibody production, as not all strains of mouse are equally
responsive to a particular antigen (Mouton et al, 1988). As SAA is a low
molecular weight antigen (Rossevatn et al, 1992) which may be too small to
118
induce immunity, an alternative strategy was employed to stimulate a response by
binding SAA to a protein carrier, apoferritin (Harlow and Lane, 1988).
However, immunisation with bovine SAA-apoferritin complex again produced no
antibody response in the sera from mice when analysed by ELISA.
Serum amyloid A was detected initially in this study using a polyclonal rabbit anti
human SAA antibody. Only a very small volume of this antibody was available
and it was not possible to obtain sufficient quantities of it from any source in
order to complete the study. An exhaustive assessment of alternative antibodies,
both monoclonal and polyclonal, failed to identify a suitable replacement antibody.
The lack of a suitable antibody for the detection of bovine SAA emphasises the
need for the production of a successful monoclonal antibody which could be used
for the accurate detection of bovine SAA in future work.
One aspect of this part of the project was to try to identify and quantify the acute
phase proteins haptoglobin and SAA in milk following intramammary challenge
with S. aureus. Haptoglobin levels were measured in the skimmed milk fraction
of samples taken from two quarters in all four cows. Our results showed an
increased haptoglobin concentration in only one quarter in one cow. The
increased haptoglobin was detected in the LH quarter which was originally
uninfected and subsequently challenged with strain B on day 8 post-challenge.
The haptoglobin concentration increased from 0.22g/l on day 3 post-challenge to
0.50g/l on day 8 post-challenge and then returned to a basal level of 0.17g/l on
day 17 post-challenge. In the other quarter of this cow, which was originally
uninfected and received no intramammary challenge, and in all quarters sampled
from the other cows, haptoglobin levels remained within the normal range despite
the presence of active infection. This suggests that haptoglobin may not be a
useful marker of inflammation in mammary secretions following infection with S.
aureus. This may be due to the chronicity of the original subclinical infection or
perhaps the challenge dose failed to induce a detectable acute phase response
when other inflammatory mediators may predominate in the gland. Another
explanation as to why elevated haptoglobin were not detected may be the choice
of days on which the samples were taken. An elevation in haptoglobin due to the
119
acute phase response may have occurred between two of the sampling dates
which were at least five days apart and returned to base levels by the time the next
samples were taken.
Due to the lack of a reliable antibody for the detection of bovine SAA, SDS-
PAGE was carried out to determine if SAA was present in milk by estimating the
molecular weight of the suspected SAA protein and by comparing it with the
semi-purified SAA. Only cow 703 was studied following analysis of haptoglobin
results, as she had been the only cow to show elevated levels of haptoglobin. A
protein of similar molecular weight to SAA was detected in the LH quarter on day
8 post-challenge. Identification of SAA by molecular weight in milk was not
straightforward as there are many other proteins in milk of similar molecular
weights to SAA, whereas serum SAA can be clearly identified by SDS-PAGE as
the majority of other serum proteins are considerably larger in size. This part of
the work is incomplete as no antibody was available to confirm immunologically
that this protein was indeed SAA at the point when the present project finished.
120
Chapter 4
General Discussion
In this study, the experimental approach was to infuse large numbers of S. aureus
bacteria into quarters which were originally infected with subclinical mastitis
caused by S. aureus, and also into quarters which were uninfected. Although the
numbers of bacteria infused during this project were much greater than in
previously published work, the S. aureus strains were derived originally from
cows with subclinical S. aureus mastitis and were probably less virulent than many
S. aureus strains. In three of the four cows studied, the resulting mastitis was
acute, with obvious swelling and reddening of the udder, but the cows did not
show signs of being sytemically ill. The fourth cow, however, became anorexic,
developed a marked pyrexia, and was subsequently treated with antibiotics.
Two main strains of S. aureus, strain A and strain B, were identified by REFP and
used in the challenge studies. Quarters which were naturally infected with S.
aureus and then challenged by the intramammary route with S. aureus, remained
persistently infected three weeks following challenge, irrespective of the challenge
strain. Quarters which were uninfected at the beginning of the project and were
subsequently challenged with either strain, also remained persistently infected
throughout the study. These results confirm the chronicity of S. aureus
intramammary infection which is well recognised in the field and in experimental
studies.
The results showed that in the cow infected subclinically with strain A,
intramammary challenge with the indigenous strain of S. aureus into the infected
quarter failed to induce an anamnestic immune response which might have
121
instigated clearance of the original infection from the quarter. In the cow infected
subclinically with strain B, intramammary challenge with the indigenous strain of
S. aureus into the infected quarter resulted in the original strain B being replaced
by a heterologous population of distinct strains R3 , R6 and R 7 in that quarter. It is
possible that a secondary immune response to the indigenous strain B resulted in
clearance of the original stimulating bacteria thus allowing a different strain to
multiply and establish infection in the quarter.
This study also suggests that intramammary challenge of subclinically infected

quarters with the non-indigenous strain did not result in permanent replacement of
the indigenous strain with the recently infused non-indigenous strain. In the cow
infected with strain A and challenged with strain B, strain A was not detected at
any point following intramammary challenge. In contrast, in the cow infected with
strain B and challenged with strain A, strain A was detected on day 3 post
challenge, although only transiently, and by day 8 the original strain B had again
become the predominant strain in that quarter.
REFP analysis also identified eight S. aureus strains other than the two main
strains involved in the study. One additional strain was designated strain A* and
was found to be very similar to strain A, with a coefficient of variation of 96%
and 24 out of 25 fragments matching on REFP analysis. Seven other additional
strains R1-R7 were also identified in the same way, but they showed lower
coefficients of variation and reduced numbers of matching fragments when
compared to strain A or strain B. It may be postulated, therefore, that while strain
A* was likely to be a clonal variant derived from strain A, the strains R 1 -R 7 may
be unrelated to strain A and strain B. The identification of these strains from
individual quarters both pre- and post-challenge indicated that a cow, or even a
quarter, may be infected simultaneously by more than one strain of S. aureus.
REFP analysis involved multiple colony analysis from each sample, which allowed
the identification of the presence of more than one S. aureus strain in an individual
quarter, and which would have failed to have been detected on routine single
colony analysis. This study showed that the examination of multiple colonies per
122
milk sample identified the diversity of S. aureus strains present and maximised the
benefit of bacterial strain identification as an epidemiological tool in mastitis
investigation. However, REFP analysis is a time consuming method, involving a
lengthy process to obtain sufficiently purified DNA. The REP-PCR technique
was developed during this project to facilitate more rapid identification of bovine
S. aureus strains. REP-PCR analysis was shown to be a successful rapid technique
for the differentiation o f all but one bovine S. aureus strain isolated in this study.
This method did not allow differentiation of strain A* from strain A. In order to
overcome this problem, digestion of the PCR product by a number of restriction
endonucleases was carried out. Of the endonucleases tested, only Stu 1 was
found to produce a variation in amplification products which enabled
discrimination of strain A* from strain A.
Future developments of this part of the work should involve further validation of
the REP-PCR with analysis of a wider range of S. aureus isolates. In this study,
the DNA was purified for REP-PCR using the method for REFP which is a two-
day process but which results in large quantities of ‘purified’ DNA. Another
development of this technique would be to improve the method for isolating
bacterial DNA, allowing a more rapid purification method to be incorporated into
the REP-PCR.
The experimental challenge model developed during this project may be more
suitable for future investigations of mammary gland infections caused by S. aureus
strains associated with subclinical mastitis, than strains derived from clinical cases
or from commonly used laboratory strains. Now that a safe, relatively repeatable,
method for infusion of these strains of S. aureus has been developed, it is
necessary for further work on intramammary challenge with different strains to be
carried out in greater numbers of cows. This is important to establish if either
changes in, or persistence of, certain S. aureus strains are the result of the strain
employed, or are the result of immune responsiveness of the individual cow. It
would be a useful avenue for further study to examine the competitive behaviour
of strains A and A* in vitro and also following intramammary challenge. It is
123
possible that strain A* is more ‘fit* than strain A in terms of persistence in the
mammary gland perhaps by virtue of reduced antigenicity.
To investigate the acute phase response as an indicator of inflammation in

mastitis, mammary secretions from the four cows with naturally occurring
subclinical mastitis caused by S. aureus were collected prior to, and following,
intramammary challenge with S. aureus. Serum amyloid A was isolated and
purified from the serum of bovine clinical cases by ultracentrifugation, gel
electrophoresis and electro-elution. Serum amyloid A was identified by enzyme
linked immunosorbent assay and immunoblotting using a polyclonal rabbit anti
human SAA antibody. The purified SAA was used to immunise mice in an
attempt to produce a monoclonal antibody against bovine SAA but this proved to
be unsuccessful, even when the SAA was bound to a carrier protein. Increased
levels of haptoglobin were identified in a single quarter of one cow. This quarter
had been originally uninfected and then challenged by the intramammary route
with strain B. In the absence of a polyclonal or monoclonal antibody to bovine
serum amyloid A, SDS-PAGE was used to identify proteins in mammary
secretions with a similar molecular weight to serum amyloid A. A protein of
similar molecular weight to serum amyloid A was identified in the same quarter in
which the increased haptoglobin levels were measured. These preliminary results
suggest that haptoglobin is not a suitable candidate as a marker of the acute phase
response in S. aureus mastitis as all but one of the infected quarters has normal
levels of this acute phase protein. Further investigation of the levels of SAA in
infected mammary secretions will require the development of suitable antibodies
or detection systems.
One possible method which could be used in the future to produce a successful
monoclonal antibody to detect bovine SAA could be to sequence and express
bovine SAA. The gene encoding bovine SAA (Rossevatn et al, 1992) would be
amplified using PCR, with the products of amplification cloned using a standard
cloning vector. Purified plasmid DNA samples could then be sequenced. The
amino acid sequence for bovine SAA has already been determined (Rossevatn et
al, 1992) and once the DNA sequence has been accurately established, a gene
124
fusion system would be employed to obtain high level expression of the SAA.
Recombinant SAA fusion proteins following purification, would be used for
monoclonal antibody production and the development of specific immunological
assays. Further work is required in order to assess the usefulness of SAA and
other acute phase proteins in integrated quality control of meat and dairy
products.
125
Glossary
1° primary
2° secondary
A* variant strain
AA amyloid A
AEC amino ethyl carbazole
APP acute phase protein (s)
APR acute phase response
B cells B lymphocytes
BCA bicinchoninic acid
BHI brain heart infusion
BSA bovine serum albumin
BMSCC bulk milk somatic cell count (s)
°C centigrade (celsius)
cm centimetre
CRP C-reactive protein
Da Dalton
dATP deoxyadenosinetriphosphate
dCTP deoxycytosinetriphosphate
dGTP deoxyguanidinetriphosphate
DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
dTTP deoxythyminetriphosphate
EDTA ethylenediaminetetra-acetate
ELISA enzyme linked immunosorbent assay
g gram
FCS fetal calf serum
HRP horseradish peroxidase
HDL high density lipoprotein
IL-1 Interleukin-1
IL-6 Interleukin-6
126
ICSCC individual cow somatic cell count (s)
IgA immunoglobulin A
IgGi immunoglobulin G type 1
IgG2 immunoglobulin G type 2
IgM immunoglobulin M
IQSCC individual quarter somatic cell count (s)
Kcl potassium chloride
kDa kilo Dalton
X lambda
1 litre
LDL low density lipoproteins
LF left fore
LH left hind
mg milligrams
MgCl magnesium chloride
ml millilitre
MLEE multilocus enzyme electrophoresis
mm millimetre
mM millimolar
NaCl sodium chloride
NAGase N-acetyl-P-D-glucosaminidase
NaH2P 0 4 sodium dihydrogen orthophosphate
Na2P 0 4 disodium hydrogen monophosphate
PBS phosphate buffered saline
PBS-T phosphate buffered saline-Tween 20
PCR polymerase chain reaction
PHA phytohaemmagglutinin
PMNL polymorphonuclear leucocyte (s)
REF restriction endonuclease fingerprinting
REFP restriction enzyme fragmentation pattern
REP-PCR repetitive extragenic palindromic polymerase chain
reaction
127
RF right fore
RH right hind
RNA ribonucleic acid
SAA serum amyloid A
SDS sodium dodecyl sulphate
SDS-PAGE sodium dodecyl sulphate polyacrylamide gel
electrophoresis
SMCC Four- (N maleimidomethyl) cyclohexane carboxylic
acid N-hydroxy succinimde ester
TBC total bacterial count (s)
TBE Tris base, boric acid disodium EDTA
TBS-T Tris buffered saline-Tween 20
T cells T lymphocytes
TE Tris base disodium EDTA
TEio lOmM Tris base, lOmM disodium EDTA
TEMED N ’, N ’, N ’, N ’,-tetramethylethylenediamine
TES Tris ethylene diamine tetra acetic acid in sodium
chloride buffer
TMN 3, 3’, 5, 5’-tetramethylbenzidine
TNFa tumour necrosis factor-alpha
Tris-HCL Tris base buffered in hydrochloric acid
TSST-1 toxic shock syndrome toxin-1
p micro
pg microgram
pi microlitre
pmol micromole
v/v volume per volume
w/v weight per volume
128
References
Adlam, C. and Easmon, C.S.F., 1983, Immunity and hypersensitivity to

staphylococcal infection. Staphylococci and staphylococcal infections, vol 1,
eds Adlam and Easmon, London, New York Academic Press, 275-323.
Aarestrup, F.M., 1995, Characterisation of Staphylococcus aureus from bovine
mastitis. A PhD Thesis.
Aarestrup, F.M., Andersen, J.K., Jensen, N.E., 1995 (a), Lack of Staphylococcal
Enterotoxin Production among Strains of Staphylococcus aureus from Bovine
Mastitis in Denmark. Acta. vet. scand, 36, 2, 273-275.
Aarestrup, F.M., Wegener, H.C., Rosdahl, V.T., 1995 (b), A Comparative Study
of Staphylococcus aureus Strains Isolated from Bovine Subclinical Mastitis
During 1952-1956 and 1992, Acta Vet. Scand. 36, 2, 237-243.
Aarestrup, F.M., Wegener, H.C., Rosdahl, V.T., 1995 (c), Evaluation of
phenotypic and genotype methods for epidemiological typing of
Staphylococcus aureus isolates from bovine mastitis in Denmark, Veterinary
Microbiology 45, 139-150.
Aarestrup, F.M., Wegener, H.C., Rosdahl, V.T., Jensen, N.E., 1995 (d),
Staphylococcal and other Bacterial Species Associated with Intramammary
Infections in Danish Dairy Herds, Acta Vet. Scand. 36, 475-487.
Aldo-Benson, M.A., Benson, M.D., 1982, SAA supppression of immune response
in vitro: Evidence for an effect on T cell-macrophage interaction. Journal of
Immunology, 126, 2390-2392.
Alsemgeest, S.P.M., Kalsbeck, H.C., Wensing, T., Koeman, J.P., Ederan, van,
A.M., Gruys, E., Ederan, van, A.M., 1994, Concentrations of serum
amyloid-A (SAA) and haptoglobin (Hp) as parameters of inflammatory
diseases in cattle. Veterinary Quarterly, 16,21-23.
Alsemgeest, S.P.M., Taverne, M.A.M., Boosman, R., van der Weydon, G.C.,
Gruys, E., 1993, Peripartum acute-phase serum amyloid-A concentration in
plasma of cows and fetuses. American Journal of Veterinary Research, 54,
164-167.
129
Anderson, J.C., 1982, Progressive pathology of staphylococcal mastitis with a note
on control, immunization and therapy. Veterinary Record 110, 372-376.
Anderson, J.C., 1986, Pathogenesis of Bacterial Infections in Animals. Library of
Congress Cataloging in Publication Data.
Arvidson, S.O., 1983, Extracellular enzymes from Staphylococcus aureus.
Staphylococcal infections, vol 12, New York Academic Press, 745-808.
Avery, O.T. and Abernethy, T.J., 1941, The occurrence during acute infections of a
protein not normally present in the blood I Distribution of the reactive protein
in patients sera and the effect of calcium on the flocculation reaction with C
polysaccharide of pneumonococcus. Journal of Experimental Medicine, 73,
173-182.
Baines, J.R., 1996, EC Hygiene Directive, A MAFF sponsored campaign to reduce
somatic cell counts. Proceedings of the British Mastitis Conference,
Stoneleigh. .
Batish, V.K., Grover, S., Sing, R.S., 1988, Thermal stability of Staphyloccocal
Enterotoxins in Milk and Related Foods. Indian Journal of Dairy Science,
497-504.
Baumann, H., Gauldie, J., 1994, The acute phase response. Immunology Today,
15, 74-80.
Belkum, A. van., 1994, DNA Fingerprinting of Medically Important Microorganisms
by Use of PCR, Clinical Microbiology Reviews, 17, 2, 174-184.
Belkum, A. van., Bax, R., Perrbooms, P., Goessens, W., van Leeuwen, N., Quint,
W.G.V., 1993, Comparison of phage typing and DNA fingerprinting by PCR
for discrimination of methicillin-resistant Staphylococcus aureus strains.
Journal of Clinical Microbiology, 31, 798-803.
Benditt, E.P., Eriksen, N., 1977, Amyloid protein SAA associated with high density
lipoprotein from human serum. Proc. Natl. Acad. Sci., 74, 4025.
Betts, J.C., Edbrooke, M.R., Thakker, R.V., Woo, P., 1991, The human acute-
phase serum amyloid A gene family: structure, evolution and expression in
hepatoma cells. Scandinavian Journal of Immunology, 34, 471-482.
Bialkowska-Hobrzanska, H., Jaskot, D., Hammerberg, O., 1990, Evaluation of
restriction endonuclease fingerprinting of chromosomal DNA and plasmid
130
profile analysis for characterisation of multiresistant coagulase-negative
staphylococci in bacteremic neonates. Journal of Clinical Microbiology, 28,
269-275.
Birgersson, A., Jonsson, P., Holmberg, O., 1992, Species identification and some
characteristics of coagulase-negative staphylococci isolated from bovine
udders, Veterinary Microbiology, 31, 181-189.
Black, 1992, Blacks Veterinary Dictionary.
Blackburn, W.D., Jr., 1994, Validity of Acute Phase Proteins as Markers of
Disease Activity, The Journal of Rheumatology, 21, 42, 9-13.
Blosser, T.H., 1979, Economic losses from and the National Research Program on
Mastitis in the United States. Journal of Dairy Science, 62, 119-127.
Blowey, R. W., 1986, An assessment of the economic benefits of a mastitis control

scheme. Veterinary Record, 119, 551-553.
Blowey, R.W., 1990, A Veterinary Book for Dairy Farmers, 2nd Edition, Farming
Press Limited.
Blowey, R.W., Andrews, A.H., Boyd, H., Eddy, R.G., 1992, Bovine Medicine-
Diseases and Husbandry of Cattle, Blackwell Scientific Publications.
Blowey, R., Edmondson, P., 1995, Mastitis Control in Dairy Herds, an illustrated
and practical guide. Farming Press Books.
Boosman, R., Niewold, Th. A., Mutsaers, Cornelius, W.A.A.M., Gruys, E., 1990,
Response of serum amyloid A concentrations in cows given endotoxin as
acute-phase stimulant. American Journal of Veterinary Research, 50, 1690-
1694.
Booth, J.M ., 1985, Bulk milk somatic cell counting: Methods in use and a proposal
for the standard presentation of data. In proceedings of the International
Dairy Federation Seminar. Progress in the Control of Bovine Mastitis, 274-
81.
Bornstein, D.L., 1982, A leukocytic pyrogen: a major mediator of the acute phase
reaction. Ann. N.Y. Acad. Sci., 389, 323-327.
B ram ley, A.J., 1978, The effect of subclinical Staphylococcus epidermidis infection
of the lactating bovine udder on its susceptibilty to infection with
131
Streptococcus agalactiae or Escherichia coli. Journal of Dairy Research, 43,
205-211.
Bramley, A.J. and Dodd, F.K., 1984, Reviews of the progress of dairy science:
mastitis control: Progress and Prospects. Journal of Dairy Research, 51, 481-
512.
Brooks, B.W., Barnum, D.A., 1984, The susceptibility of bovine udder quarters
colonised with Corynebacterium bovis to experimental infection with
Staphylococcus aureus or Streptococcus agalactiae. Canadian Journal of
Complementary Medicine, 48, 146-150.
Brown, T.A., 1991, Essential Molecular Biology, Vol. II, A practical approach,
Oxford University press, New York, 57-209.
Burnie, J.P., Lee, W., 1988, A comparison of DNA and immunoblot fingerprinting
of the SII biotype of coagulase negative staphylococci. Epidemiology and
Infection, 101, 203-212.
Butler, J.E., 1986, Biochemistry and biology of ruminant immunoglobulins.
Progress in Veterinary Microbiology and Immunology, 2, 1-53.
Casl, M.T., Grubb, A., 1993, A rapid enzyme-linked immunosorbent assay for
serum amyloid A using sequence-specific antibodies. Annals of Clinical
Biochemistry, 30, 278-286.
Castle, M.E., 1985, Hygienic Milk Production, The Scottish Milk Marketing Board.
Christ, D., Schwarz, S., Lammler, C., 1988, DNA fingerprinting of Streptococcus
uberis. Medical Scientific Research, 1297-1298.
Colditz, I.G., Watson, D.L., 1985, The immunophysiological basis for vaccinating
ruminants against mastitis. Australian Veterinary Journal, 62, 145-153.
Collins R.A., Parsons. K.R., Bland, A.P., 1986, Antibody-containing cells and
specialised epithelial cells in the bovine teat. Research in Veterinary Science,
41, 50-55.
Concha, C., Holmberg, O., 1990, Ability of bovine mammary macrophages to
enhance proliferation of autologous blood and mammary secretion
lymphocytes. Journal of Dairy Research, 57, 7-16.
Concha, C., Holmberg, O., Morein, B., 1978, Proportion of B- and T- lymphocytes
in normal milk. Journal of Dairy Research, 45, 287-290.
132
Concha, C., Holmberg, O., Morein, B., 1980, Characterisation and response to
mitogens of mammary lymphocytes from the bovine dry period. Journal of
Dairy Research, 47, 305-311.
Conner, J.G., Eckersall, P.D., Doherty, M., Douglas, T.A., 1986, Acute phase
response and mastitis in the cow, Research in Veterinary Science 41, 126-
128.
Conner, J.G., Eckersall, P.D., Wiseman, A., Aitchison, T.C., Douglas, T.A.,
1988, Bovine acute phase response following turpentine injection. Research
in Veterinary Science, 44, 82-88.
Counter, D.E., 1981, Outbreak of bovine mastitis associated with Corynebacterium
bovis. Veterinary Record, 108,560-561.
Craven, N., 1987, Efficacy and financial value of antibiotic treatment of bovine
clinical mastitis during lactation- a review. Brittish Veterinary Journal, 143,
410-421.
Craven, N. and Williams, M.R., 1985, Defences of the bovine mammary gland
against infection and prospects for their enhancement. Veterinary
Immunology Immunopathology, 10, 71-77.
DeGraves, F.J., Fetrow, J., 1993, Economics of mastitis and mastitis control.
Veterinary Clin. North America (Food Animal Pract.), 9, 421-434.
Devriese, L.A., 1984, A simplified system for biotyping Staphylococcus aureus
strains isolated from different animal species, Journal of Applied
Bacteriology 56, 215-220.
Dice, L.R., 1945, Measures of the amount of ecological association between species.
Ecology, 26, 297-302.
Duncan, J.R,, Wilkie, B.N., Hiestand, F., 1972, The serum and secretory
immunoglobulins of cattle: characterisation and quantitation. Journal of
Immunology, 108, 965-976.
Eckersall, P.D., Personal Communication.
Eckersall, P.D., 1992 (a), Animal Acute Phase Protein: Past, Present and Future, Vth
Congress of ISACB, 97-103.
Eckersall, P.D., 1992 (b), Meat Inspection - The Potential of Acute Phase Protein
Assay, Meat Focus International, 279-282.
133
Eckersall, P.D., 1988, Protein Analyses and Veterinary Diagnosis in Animal
Clinical Biochemistry-The Future. British Library Cataloguing in Publication
Data.
Eckersall, P.D., 1995, Current Concepts on the use of Acute Phase Proteins in
Clinical Veterinary Medicine, Animal Biochemistry 32, 2, 69-72.
Eckersall, P.D., Conner, J.D., 1988, Bovine and Canine Acute Phase Proteins,
Veterinary Research Communications, 12, 169-178.
Eisenstein, B.I., 1990, New molecular techniques for microbial epidemiology and the
diagnosis of infectious diseases. Journal of Infectious Disease, 161, 595-
602.
Emanuelson, U., Funke, H., 1991, Effect of milk-yield on relationships between
bulk milk somatic cell count and prevalence of mastitis. Journal of Dairy
Science, 74, 2479-2483.
Emery, P., Luqmani, R., 1993, The Validity of surrogate markers in rheumatic
disease. British Journal of Rheumatology, Supplement 3, 32, 3-8.
Esselmont, R.J. and Spincer, I., 1992, The Incidence and Costs of Diseases in
Dairy Herds Daisy- The Dairy Information System, Report No 2.
Evans, S.W. and Whicher, J.T., 1992, Biochemistry of Inflammation, Immunology
and Medicine, Kluwer Academic Publishers.
Fitzpatrick, J.L., Williams, N.A., Bailey, M., Bland, P.W., Stokes, C.R., 1995,
Presentation of soluble and bacterial antigens by milk-derived cells to
unprimed bovine T cells in vitro. Veterinary Immunology and
Immunopathology, 46, 251-266.
Foster, T.J., 1991, Potential for vaccination against infections caused by
Staphylococcus aureus, Vaccine, 9, 221-227.
Fox, L.K., Gay, J.M., 1993, Contagious Mastitis Update on Bovine Mastitis.
Veterinary Clinics of North America: Food and Animal Practise, 9, 475-487.
Fox, L.K., Hutton, C.T., Hancock, D.D., Gersham, M., 1990, Typing
Staphylococcus aureus to determine sources of intramamammary infection
(IMI). Journal of Dairy Science, 73, 1, 257.
134
Gilson, E.J.M ., Clement, M., Britlag, D., Hofnung, M., 1989, A family of
dispersed repetitive, extragenic palindromic DNA sequences in E. coli.
EMBOJ, 3, 1417-21.
Godenir, N.L., Jeenah, M.S., Coetzee, G.A., Westhuyzen, D.R.V., Strchan, A.F.,
De Beer, F.C., 1985, Standardisation of the quantitation of serum amyloid A
protien (SAA) in human serum. Journal of Immunological Methods, 83, 217-
225.
Goetzl, E.J., 1980, A role for endogenous mono-hydroxy-eicosatetraenoic acids
(HETEs) in the regulation of human neutrophil migration. Immunology, 49,
709-719.
Greenberg, D.P., W ard, J.I., Bayer, A.S., 1987, Influence of Staphylococcus
aureus antibody on experimental endocarditis in rabbits. Infection and
Immunity, 55, 3030-3034.
Gruys, E., Ederen, A.M., Alsemgeest, S.P.M., Kalsbeek, H.C., Wensing, T.,
1993, Acute phase protein values in blood of cattle as indicator of animals
with pathological processes. Archiv fur Lebensmitteihygien, 44, 105-128.
Gruys, E., Obwolo, M .J., Matilda, J.M ., Toussaint, M .J.M ., 1994, Diagnostic
significance of the major acute phase proteins in veterinary clinical
chemistry: a review, Veterinary Bulletin 64, 11, 1009-1018.
Guidry, A.J., Butler, J.E., Pearso, R.E., Weinland, B.T., 1980, IgA, IgGi, IgG2
and BSA in serum and mammary secretion throughout lactation. Veterinary
Immunology and Immunopathology, 1, 329-341.
Guidry, A.J., Berning, L.M., Hambleton, C.N., 1993, Opsonisation of
Staphylococcus aureus by bovine immunoglobulin isotypes. Journal of Dairy
Science, 76, 1285-1289.
Guinee, P.A.M., van Leeuwen, W .J., 1978, Phage typing of Salmonella. Methods
in Microbiology, 11, 157-191.
Harlow, E., Lane, D., 1988, Antibodies, A Laboratory Manual, Cold Spring Harbour
Limited.
Harvey, J., Gilmour, A., 1990, Isolation and identification of staphylococci from
milk powders produced in Northern Ireland. Journal of Applied
Bacteriology, 68, 433-438.
135
Hibbitt, K.G., Benians, M., 1971, Some effects in vivo of the teat canal and effects
in vitro of cationic proteins on staphylococci. Journal of General
Microbiology, 68, 123-128.
Higgins, C.F., Ames, G.F.L., Barnes, W.M., Clement, J.M ., Hofnung, M., 1982,
A novel intercistronic regulatory element of prokaryotic operons. Nature,
298, 760-762.
Hill, A.W., 1981, Factors influencing the outcome of Escherichia coli mastitis in the
dairy cow. Research in Veterinary Science, 31, 107-112.
Hill, A.W. and Leigh, J.A., 1989, DNA fingerprinting of Streptococcus uberis: a
useful tool for epidemiology of bovine mastitis. Epidemiology and Infection,
103, 165-171.
Hillerton, J.E., Bramley, A.J., Staker, R.T. and McKinnon, C.H., 1995, Patterns
of intramammary infection and clinical mastitis over a 5 year period in a
closely monitored herd applying mastitis control measures. Journal of Dairy
Research, 62, 39-50.
Hirvonen, J., Pyorala, S., Jousimies-Somer,H., 1996, Acute phase response in
heifers with experimentally induced mastitis. Journal of Dairy Research, 63,
351-360.
Hoffman, J.S., Benditt, E.P., 1982, Changes in high-density lipoprotein content
following endotoxin administraion in the mouse. Formation of Serum
amyloid protein rich sub-fractions. Journal of Biological Chemistry, 257,
10510-10517.
Hogan, J.S., White, D.G. and Pankey, J.W., 1987 Effects of teat dipping on
intramammary infections by Staphylococci other than Staphylococcus aureus.
Journal of Dairy Science, 70, 873-879.
Hoi, P.R., Gruys, E., 1984, Amyloid A proteins in different species. Applied
Pathology, 2, 316-327.
Hoi, P.R., Snel, F.W .J.J., Draaijer, M., Gruys, E., 1987, The serum amyloid A
stimulating factor (SAASF) in the hamster. Journal of Comparitive
Pathology, 97, 677-685.
Horadogoda, A., 1994, Serum amyloid A and the acute phase response in bovine
pneumonic pasteurellosis. A PhD Thesis.
136
Horadagoda, A., Eckersall, P.D., Alsemgeest, S.P.M., Gibbs, H.A., 1993,
Purification and quantitative measurement of bovine serum amyloid-A.
Research in Veterinary Science 55, 317-325.
Horadogoda, A., Eckersall, P.D., Hodgson, J.C., Gibbs, H.A., Moon, G.M.,
1994, Immediate responses in serum TNF alpha and acute phase protein
concentrations to infection with Pasteurella haemolytica AI in calves.
Research in Veterinary Science, 57, 129-132.
Honkanen-Buzalski, T., Katila, T., Sandholm, M., 1981, Milk antitrypsin activity
durimg clinical and experimental bovine mastitis. Acta. vet. scand., 22, 360-
368.
Hurley, D.J., Kensinger, M.H., Mastro, A.M., Wilson, R.A.,1990, An evaluation
of the mononuclear cells derived from bovine mammary gland secretions
using leucocyte antigen specific monoclonal antibodies, light scattering
properties and non-specific esterase staining. Veterinary Immunolgy and
Immunopathology, 25, 177-193.
Husby, G., Natvig, J.B., 1974, A serum component related to nonimmunoglobulin
amyloid protein AS, as possible precursor of the fibrils. Journal of Clinical
Investigation, 53, 1054-1061.
Huszenicza, G. and Stollar, Z., 1993, Possibilities of the diagnosis of subclinical
mastitis. Review article. Magyar-Allatarvosok-Lapja, 48, 8, 460-465.
Janeway-Travers, 1994, Immunobiology. The Immune System in health and
Disease. Blackwell Scientific Publications Oxford.
Jayarao, B.M., Oliver, S.P., Tagg, J.R., Matthews, K.R., 1991, Genotypic and
phenotypic analysis of Streptococcus uberis isolated from bovine mammary
secretions. Epidemiology and Infection, 107, 543-555.
Jayarao, B.M., Bassam, B.J., Caetario-Anolles, G. Gresshoff, P.M., Oliver, S.P.,
1992, Subtyping of Streptococcus uberis by DNA Amplification
Fingerprinting. Journal of Clinical Microbiology, 30, 5, 1347-1350.
Jonsonn, P., Holmberg, O., 1981, Staphylococcus aureus isolated from acute
and chronic mastitis. Comparitive studies on bacterial surface
characteristics, biochemical properties and phage typing. In
Staphylococci and Staphylococcal infections, 793-801.
137
K apur, V., Sischo, W.M., Greer, R.S., W hittam, T.S., Musser, J.M ., 1995,
Molecular Population Genetic Analysis of Staphylococcus aureus Recovered
from Cows. Journal of Clinical Microbiology, 33, 22, 376-380.
Kennedy, M.W., Qureshi, F., Haswell-Elkins,M. and Elkins, D.B., 1987,
Homology and heterology between the secreted antigens of the parasitic laral
stages of Ascaris lumbricoides and Ascaris suum. Clinical Experimental
Kenny, K., Reiser, R.F., Bastida-Corcuera, F.D., Norcross, N.L., 1993,
Production of Enterotoxins and Toxic Shock Syndrome Toxin by Bovine
Mammary Isolates of Staphylococcus aureus. Journal of Clinical
Microbiology, 706-707.
Kent, J., 1992, Acute phase proteins: Their Use in Veterinary Diagnosis, British
Veterinary Journal 148, 279-281.
Kisilevsky, R , 1991, Serum amyloid A (SAA), a protein without a function: Some
suggestions with reference to cholesterol metabolism. Medical Hypotheses,
35,337-341.
Kloos, W.E. and Bannerman, T.L., 1994, Update on clinical significance of
coagulase negative staphylococci. Clinical Microbiology Review, 7, 117-140.
Kloos, W.E., Schleifer, K.H., 1986, Staphylococcus. Bergey’s Manual of Systemic
Bacteriology, vol 2. Wiiliams and Wilkins, 1013-1035.
Koj, A., 1974, Structure and Function of Plasma Proteins. Ed. A.C. Alison, Plenum
Publishing, 73-131.
Kotzin, B.L., Leung, D.Y., Kappler, J., M arrack, P., 1993, Superantigens and
their potential role in human disease. Advances in Immunology, 54, 99-166.
Kremer, W.D.J., Noordhuizen-Stassen, E.N., Lohuis, J.A.C.M., 1990, Host
Defences and Bovine Coliform Mastitis. Vet. Quarterly, 12, 103-113.
Kushner, L, 1982, The phenomenon of the acute phase response. Ann. N.Y. Acad.
Sci., 389, 39-48.
Kushner, I., Broder, M.L., K arp, D., 1978, Control of the acute phase response.
Serum C-reactive protein kinetics after myocardial infarction. Journal of
Clinical Investigation, 61, 235-242.
138
Kushner, I., Mackiewicz, A., 1987, Acute phase proteins as disease markers.
Disease Markers, 5, 1-11.
Lacy-Hubbert S.J., Hillerton, J.E., 1995, Physical characteristics of the bovine teat
canal and their susceptibility to streptococcal infection. Journal of Dairy
Research, 62, 395-404.
Lam, T.J.G.M., 1996, Dynamics of Bovine Mastitis. A field study in Low Somatic
Cell Count Herds.
Lam, T.J.G.M., Lipman, L.J.A., Shukken, Y.H. Gaastra, W., Brand, A., 1996,
Epidemiological characteristics of bovine clinical mastitis caused by
Escherichia coli and Staphylococcus aureus studied by DNA fingerprinting,
American Journal of Veterinary Research 57, 39-42.
Laurell, C.B., 1985, Recent Advances in Clinical Biochemistry No.3. Eds.
K.G.M.M. Alberti, C.P.Price, Churchill Livingston.
Lee, C.S., Wooding, F.B.P., Kemp, P., 1980, Identification, properties, and
differential counts of cell populations using electron microscopy of dry cows
secretions, colostrum and milk from normal cows. Journal of Dairy
Research, 47, 39-50.
Lee, J.C., 1996. The prospects for developing a vaccine against Staphylococcus
aureus (Review). Trends in Microbiology, 4, 162-166.
Levin, M., Pras, M., Franklin, E.C., 1973, Immunologic studies of the major
nonimmunoglobulin protein of the amyloid I. Identification and partial
characterisation of a related serum component. Journal of Experimental
Medicine, 138, 373-380.
Linde, C., Holmberg, O., Astrom, G., 1975, Interference between Staphylococcus
epidermidis and Staphylococcus aureus in the bovine udder. Acta. Vet.
Scand., 16, 146-148.
Linde, C., Holmberg, O., Astrom, G.,1980, The interference between coagulase
negative staphylococci and Corynehacterium bovis and the common udder
pathogens in the lactating cow. Nord. Vet. Med., 32, 552-558.
Lipman, L.J.A., Nijs, A.de., Lam, T.J.G.M., Gaastra, W., 1995, Identification of
Escherichia coli strains from cows with clinical mastitis by serotyping and
DNA polymorphism patterns with REP and ERIC primers. Veterinary
Microbiology, 43, 13-19.
139
Lipman, L.J.A., Nijs, A.D., Lam, T.J.G.M., Rost, J.A., Dijk, L.V., Schukken,
Y.H., Gaastra, W., 1996, Genotyping by PCR, of Staphylococcus aureus
strains, isolated from mammary glands of cows. Veterinary Microbiology 48
51-55.
Logue, D.N., Gunn, J., Chaplin, S., 1995, Milk Quality, Mastitis and Somatic Cell
Counts controlling your bulk tank somatic cell count. Proceedings of the
British Mastitis Conference, Stoneleigh.
Logue and Hillerton, J.E., 1996, Biotechnology and Biological Sciences Research
Council Agriculture Systems Directorate, Dairy cows welfare management,
Report of work shop held 30th April- 1st May 1996.
Lupski, J.R., Weinstock, G.M., 1992, Short, interspersed repetitive DNA sequences
in prokaryotic genomes. Journal of Bacteriology, 174, 4525-4529.
M adders, W.M., 1995, Impact of EU directives on the UK Dairy Farmer.
Proceedings of the British Mastitis Conference, Stoneleigh.
Madgwick, L., Mayer, S., Keen, P., 1989, Penetration of antibiotics into bovine
neutrophils and their activity against intracellular Staphylococcus aureus.
Journal of Antimicrobial Chemotherapy, 24, 709-718.
M akaya, P.V., 1996, Distribution of common mastitis pathogens and the use of
bacteriological typing methods as epidemiological tools in mastitis control
(Msc. Thesis). Department of Veterinary Microbiology, The Royal
Veterinary and Agricultural University, Copenhagan, Denmark.
M akim ura, S. and Suzuki, N,. 1982, Quantitative Determination of Bovine Serum
Haptoglobin and Its Elevation in Some Inflammatory Diseases., Japanese
Journal of Veterinary. Science 44 15-21.
Marinkovic, S., Jahreis, G.P., Wong, G.G., Baumann, H., 1989, IL-6 modulates
the synthesis of a specific set of acute phase proteins in vivo. Journal of
M atsunaga, T., Kam ata, S., Kakiichi, N., Uchida, K., 1993, Characteristics of
Staphylococcus aureus Isolated from Peracute, Acute and Chronic Bovine
Mastitis, Journal of Veterinary Medical Science, 55, 2, 297-300.
Matthews, K.R., Jayarao, B.M., Oliver, S.P., 1992, Restriction endonuclease
fingerprinting of genomic DNA of Staphylococcus species of bovine origin.
Epidemiology and Infection, 109, 59-68.
140
Matthews, K.R., Kum ar, S.J., O’connor S.A., Harmon, R.J., Pankey, J.W ., Fox,
L.K., Oliver, S.P., 1994, Genomic fingerprints of Staphylococcus aureus of
bovine origin by polymerase chain reaction-based DNA fingerprinting.
Epidemiology and Infection 112, 177-186.
Maudsley, S., 1985, PhD Thesis, University of London.
M aurhaug, G.,1983, Three assays for the charactreisation and quantitation of human
serum amyloid A. Scandinavian Journal of Immunology, 18, 329-338.
M aurhaug G. and Dowton, S.B. 1984, Serum amyloid A: An Acute phase
apolipoprotein and precursor of AA amyloid., Baillieres Clinical
Rheumatology 8, 3, 553-573.
McDonald, J.S., Anderson, A.J., 1981, Antibiotic sensitivity of S. aureus and CNS
isolated from infected bovine mammary glands. Cornell Vet., 71, 391-396.
McDonald, J.S., 1970, Microscopic observations of teat canals from susceptible and
resistant bovine mammary glands: a preliminary report. Proceedings, 6th
International Conference on Cattle Disease, 97-103.
McDonald, J.S., 1975, Radiographic method for anatomic study of the teat canal:
changes between milking periods. American Journal of Veterinary Research,
36, 1241-42.
McGuire, T.C., Mosoke, A.J., K urtti, T., 1979, Functional properties of bovine
IgGi and IgG2 : interaction with complement, macrophages, neutrophils and
skin. Immunology, 38,249-256.
McKeand, J., 1992, A PhD Thesis
McKellar, Q.A., 1996, How antimicrobials work in mastitis. Proceedings of the
British mastitis conference, Stoneleigh.
Menzies, F.D., Bryson, D.G., McCallion, T., Matthews, D.I., 1995, A study of
mortality among suckler and dairy cows in Northern Ireland in 1992.
Veterinary. Record, 137, 531-536.
Miller.R.H., Pankey, J.W ., Bright, S.A., Bitman, J., Wood, D.L., Capuco, A.V.,
1991, Proceedings of the USD A, Agricultural Research Service, University of
Vermont, Burlington.
141
Mori, M., Hayward, A.R., 1982, Phenotype and function of human milk monocytes
as antigen presenting cells. Clinical Immunology and Immunopathology, 23,
94-99.
M orimatsu, M., Sarikaputi, M., Syuto, B., Saito, M., Yamamoto, S., Naiki, M.,
1992, Bovine haptoglobin: single radial immunodiffusion assay of its
polymeric forms and dramatic rise in acute-phase sera., Veterinary
Immunology and Immunopathology, 33, 365-372.
Mouton, D., Sant’ anna-O.A., Biozzi, G., 1988, Multigenic control of specific and
non-specific immunity in mice. A review. Livestock Production Science, 20,
277-286.
Myllys, V., 1995, Staphylococcal mastitis in heifers and dairy cows. PhD Thesis,
National Veterinary and Food Research Institutue.
Neave, F.K., Dodd, F.H., Kingwill, R.G., W estgarth, D.R., 1969, Control of
mastitis in the dairy herd by hygiene and management. Journal of Dairy
Science, 52, 696-707.
Newman, S.S., Roy, A.S., Luther, D.G., Hoskins, J.D., Cox, H.U., Kearney,
M.T., 1985, Comparitive evaluation of microdilation method and
conventional disk diffusion method for determination of antibiograms of
Gram +ve cocci. American Journal of Veterinary Research, 46, 294-297.
Nickerson, S.C., 1985, Immune mechanisms of the bovine udder: An overview.
Journal of American Veterinary Medicine Ass., 187, 1, 41-45.
Nickerson, S.C., Boddie, R.L., 1994, Effect of naturally occurring coagulase-
negative staphylococcal infections on experimental challenge with major
mastitis pathogens. Journal of Dairy Science, 77, 2526-2536.
Nickerson, S.C., Heald, C.W., 1982, Cells in local reactions to experimental
Staphylococcus aureus infection in bovine mammary gland. Journal of Dairy
Science, 65, 105-116.
Niewold, Th.A., Tooten, P.C.J., 1991, Purification and characterisation of hamster
serum amyloid A protein (SAA) by cholesterol hemisuccinate affinity
chromatography. Scandinavian Journal of Immunology, 31, 942-949.
Nizamlioglu, M., Tekeli, T., Erganis, O., Baspinar, N., 1989, Biochemical and
microbiological studies on subclinical bovine mastitis. Veteriner-Fakultesi-
Dergisi, 5, 135-143.
142
Norcross. N.L., Burvenisch, C., Vandeputte-Van-Messom, G., Hill, A.W., 1991,
New insight into the pathogenisis of mastitis. Rijkuniver-Siteit Gent, 129-
139.
Novick, R.P., 1990, Molecular Biology of the Staphylococci, Library of Congress
Cataloging in Publication Data.
Oeding, P., 1983, Taxonomy and Identification. ‘Staphylococci and Staphylococcal
infections’ eds. C.S.F.Easmon, C.Adlam, Academic Press Inc.
Oliver, S.P., Lewis, M. J., Ingle, T.L., Gillespie, B.E., Matthews, K.R., Dowlen,
H.H., 1993, Premilking teat disinfection for the prevention of environmental
pathogen intramammary infections. Journal of Food Protect., 56, 852-855.
Olsen, I.E ., Brown, V.J., Skove, M.N., Christensen, J.P., 1993, Bacterial Typing
Methods suitable for Epidemiologic analysis- Applications in Investigations
of Salmonelossisi among Livestock. The Veterinary Quarterly, vol, 15, no 4,
125-135
Orden, J.A., Goyach, J., Hernandez, J., Domenech, A., Suarez, G., Gomez-
Lucia, E., 1992, Detection of enterotoxins and TSST-1 secreted by
Staphylococcus aureus isolated from ruminant mastitis. Comparison of
ELISA and immunoblot. Journal of Applied Bacteriology, 72, 486-489.
Outteridge, P.M., Lee, C.S., 1981, Cellular immunity in the mammary gland with
particular reference to T, B lymphocytes and macrophages. Adv. Exp. Med.
Biol., 137,513-534.
Outteridge, P.M., and Lee, C.S., 1988, The defence mechanisms of the mammary
gland of domestic ruminants. Progress Veterinary Microbiology Immunology
4, 165-169.
Paape, M .J., Corlett, N.J., 1984, Intensification of milk somatic cell response to
intramammary device. American Journal of Veterinary Reseasrch, 45, 1572-
1575.
Paape M .J., Guidry, A.J., Kirk, S.T., Bolt, D.J., 1975, Measurement of
phagocytosis of 32P -labeled Staphylococcus aureus by bovine leukocytes:
Lysostaphin digestion and inhibitory effect of cream. American Journal of
Veterinary Research, 36, 1737-1743.
143
Paape, M .J., Guidry, A.J., 1977, Effect of fat and casein on intracellular killing of
Staphylococcus aureus by milk leucocytes. Proc. Soc. Exp. Biol. Med., 155,
588.
Pankey, J.W ., Boddie, N.T., W atts, J.L., Nickerson, S.C., 1985, Evaluation of
protein A and commercial bacterin as vaccines against Staphylococcus aureus
mastitis by experimental challenge. Journal of Dairy Science, 68, 726-731.
Pankey, J.W ., Wildman, E.E., Drechsler, P.A., Hogan, J.S., 1987, Field trial
evaluation of premilking teat disinfection. Journal of Dairy Science, 70, 867-
872.
Pearce, C., 1995, What the purchaser requires in quality from the farmer.
Proceedings of the British Mastitis Conference, Stoneleigh.
Pepys, M.B., 1979, Acute phase phenomena. In Rheumatology and Immunology,
eds A.S. Cohen, 85-88.
Pepys, M.B., and Baltz, M.L., 1983, Acute Phase Proteins with Special Reference
to C-Reative Protein and Related Proteins (Pentaxins) and Serum Amyloid A
Protein. Advances in Immunology, 34, 141-212.
Pepys, M.B., Baltz, M.L., Tennent, G.A., Kent, J., Ousey, J., Rossdale, P.D.,
1989, Serum amyloid A protein (SAA) in horses: objective measurement of
the acute phase response. Equine Veterinary Journal, 21, 106-109.
Platt, D.J., Sullivan, L., 1992, Molecular Epidemiolgy: Molmatch and Beyond.
International Labmate XVI Issue VII.
Platt, D.J., Candlish, D., Brown, A., Gillies, E., Edwards, G., 1994, Strategic
application of molecular methods in outbreak investigation. Journal of
Medical Microbiology, 41, 271.
Plikaytis, B.D., Carlone, G.M., Edmonds, P., Mayer, L.W., 1986, Robust
Estimation of Standard Curves for Protein Molecular Weight and Linear-
Duplex DNA Base-Pair number after Gel Electrophoresis. Analytical
Biochemistry, 152, 346-364.
Pott, M .J., 1996, Mastitis Therapy-What I am trying to achieve. Proceedings of the
Prescott, L.M., Harley, J.P., Klein, D.A., 1990, Microbiology, William C. Brown
Publishers.
144
Prescott, L.M., Harley, J.P., Klein, D.A., 1996, Microbiology, Volumes 1 and 2,
William C.Brown Publishers.
Pyorala, S., 1988, Indicators of inflammation to evaluate the recovery from acute
bovine mastitis. Research in Veterinary Science, 45, 166-169.
Radostits, O.M., Blood, D.C., Gay, C.C., 1994, Veterinary Medicine: A Textbook
of the Diseases of Cattle, Sheep, Pigs, Goats and Horses. Eighth Edition,
Balliere Tindall.
Rantamaki, L.K., and Muller, H.P., 1992, Isolation and characterization of a2
macroglobulin from mastitis milk. Journal of Dairy Research, 59, 273-285.
Rantamaki, L.K., and Muller, H.P., 1995, Phenotypic characterization of
Streptococcus dysgalactiae isolates from bovine mastitis by their binding to
host derived proteins. Veterinary Microbiology 46, 415-426.
Raynes, J.G., McAdam, K.P.W .J., 1988, Purification of serum amyloid A and other
high density apolipoproteins by hydrophobic interaction chromatography.
Analytical Biochemistry, 174, 116-124.
Rebhun, W.C. 1995, Diseases of Dairy Cattle, Williams and Wilkins, Library of
Congress Cataloging in Publication.
Reiter, B., 1978, Review of the progress of Dairy Science: antimicrobial systems in
milk. Journal of Dairy Research, 45, 131-147.
Roberson, J.R., Fox, L.K., Hancock, D.D., Gay, J.M ., 1994, Ecology of
Staphylococcus aureus, Isolated from Various Sites on Dairy Farms. Journal
of Dairy Science, 3354-64.
Roitt, L, Brostoff, J., Male, D., 1993, Immunology, 3rd Edition, Mosby-Year book
Europe Limited.
Roitt, I.M. and Delves, P.J., 1992, Encyclopedia of Immunology, Academic Press.
Rossevatn, K., Andresen, P.K., Sletten, K., Husebekk, A., Husby, G.,
Nordstoga, K., Johnson. K.H., Westermark, G.T., W estermark, P.,
1992, The Complete Amino Acid Sequence of Bovine Serum Amyloid
Protein A (S AA) and of Subspecies of the Tissue-Deposited Amyloid Fibril
Protein A , Scandanavian Journal of Immunology, 35, 217 - 224.
Russell, M.W., Reiter, B., 1975, Phagocytic deficiency of bovine milk leucocytes:
and effect of casein. Journal Reticulo. Soc., 18, 1-13.
145
Saini, P.K., Webert, D.W., 1991, Application of acute phase reactants during
antemortem and postmortem meat inspection. Journal of the American
Veterinary Medical Association, 198, 1898-1901.
Salonen, M., Hirvonen, J., Pyorala, S., Sankari, S., Sandholm, M., 1996,
Quantitiative determination of bovine serum haptoglobin in experimentally
induced E. coli mastitis. Research in Veterinary Science, 60, 88-91.
Sandholm, M., Honkanen-Buzalski, T., Kangasnieli, R , 1984, Milk
trypsin-inhibitor capacity as an indicator of bovine mastitis-a novel principle
which can be automated. Journal of Dairy Research, 51, 1-9.
Sandholm, M., and M attila, T., 1986, Biochemical Aspects of Bovine Mastitis. Isr.
Journal of Veterinary Medicine, 42, 4, 405-415.
Saulnier, P., Bourneix, C., Prevost, G., Andremont, A., 1993, Random amplified
polymorphic DNA assay is less discrimant than pulse-field gell elctrophoresis
for typing strains of methicillin-resistant Staphyloccocus aureus. Journal of
Clinical Microbiology, 31, 982-985.
Schlam, O.W., Smith, R , Kaneko, J.J., 1970, Plasma protein: fibrinogen ratios in
dogs, cattle and horses. Influence of age on normal values and explanation of
use in disease. California Vet., 24, 9-11.
Schroldt, W., Kruger, M., Hien, T.T., Fuldner, M., Kunze, R , 1995. C-reactive
protein as a parameter of mastitis. Tierarztl-Prax, 23, 337-41.
Schukken, Y.H., Van de Geer, D., Grommers, F.J., Smit, J.A.H., Brand, A.,
1989, Intramammary infections and risk factors for clinical mastitis in herds
with low somatic cell counts in bulk milk. The Veterinary Record, 125, 393-
396.
Sears, P.M., Gonzalez, R N ., Wilson, D.J., Han, H .R , 1993, Procedures for
mastitis diagnosis and control. Veterinary Clinics North America (Food
Animal Pract.), 9, 445-468.
Sears, P.M., Heider, L.E., 1981, Detection of Mastitis-Symposium on Herd Health
Management- Dairy Cow. Veterinary Clinics of North America: Large
Animal Practice, 3, 327-346.
Sears, P.M., Smith, B.S., English, P.B., Herer, P.S., Gonzalaz, R N ., 1990,
Shedding Pattern of Staphylococcus aureus from Bovine Intramammary
Infections. Journal of Dairy Science, 73, 2785-2789.
146
Sears, P.M., Wilson, D.J., Gonzalez, R.N., Hancock, D.D., 1991, Microbiological
Results from Milk Samples Obtained Premilking and Postmilking for the
Diagnosis of Bovine Intramammary Infections. Journal of Dairy Science, 74,
4183-4188.
Shainkin-Kestenbaum, R., Berlyne, G., Zimlichman, S., Sorin, H.R., Nyska, M.,
Danon, A.r , 1991, Acute phase protein, serum amyloid A, inhibits IL-1 and
TNF-induced fever and hypothalamic PGE2 in mice. Scandinavian Journal of
Shepers, J., Lam, T.J.G.M., Schukken, Y.H., Wilmink, J.B.M. and Hanekamp,
W.J.A., 1996, Development and evaluation of quarter somatic cell count
thresholds (submitted for publication)
Shuster, D.E., Harmon, R.J., Jackson, J.A., Heniken, R.W., 1991, Suppression of
Milk Production During Endotoxin-Induced Mastitis. Journal of Dairy
Science, 74, 3763-3774.
Shuster, D.E., Lee, E.K., Kehrli, M.E., 1996, Bacterial growth, inflammatory
cytokine production, and neutrophil recruitment during coliform mastitis in
cows within ten days after calving, compared with cows at midlactation.
American Journal of Veterinary Research, 57, 1569-1575.
Sipe, J.D., Gonnerman, W.A., Loose, L.D., Knapschaefer, G., Xie, W .J.,
Franzblau, C., 1989, Direct binding enzyme linked immunosorbent assay
(ELISA) for serum amyloid A (SAA). Journal of Immunological Methods,
125, 125-135.
Skinner, J.G., Brown, R.A., Roberts, L. 1991, Bovine haptoglobin response in
clinically defined field conditions., Veterinary Record, 128, 147-149.
Slanetz, L.W., Bartley, C.H., Allen, F.E., 1963, Vaccination of Dairy Cattle
Against Staphylococcic Mastitis. American Journal of Veterinary Research,
24, 102, 923-931.
Smith, C.W., Goldman, A.S., 1968, The cells of human colostrum. I. In vitro
studies of morphology and function. Pediatric Research, 2, 103-109.
Smith, K.L., Hogan, J.S., 1995, The importance of coagulase-negative
staphylococci. Int. Dairy Federation Mastitis Newsletter, 142,26-29.
Smith, J., Shultz, R., 1977, Mitogen- and antigen-responsive milk lymphocytes.
Cellular Immunology, 29, 94-99.
147
Smith, K.L., Todhunter, D.A., Schoenberger, P.S., 1985, Symposium:
Environmental Effects on Cow Health and Performance. Environmental
mastitis: Cause, Prevalence, Prevention. Journal of Dairy Science, 68, 1531-
1553.
Sumner, J., 1994, A view of market demands on dairy farmers. Proceedings of the
Sutra, L., and Poutrel, B., 1993, Virulence factors involved in the pathogenesis of
bovine intramammary infections due to Staphylococcus aureus. Journal of
Medical Microbiology, 40, 79-89.
Tobias, P.S., McAdam, K.P.W .J., Ulevitch, R.J., 1982, Interactions of bacteria
lipopolysaccharide with acute phase serum and isolation of two forms of
rabbit serum amyloid A. Journal of Immunology, 128, 1420-7.
Tomlinson, L.A., Christie, J.F., Fraser, E.M., McLaughlin, D., McIntosh, A.E.
and Kennedy, M.W., 1989, MHC Restriction of the antibody repertoire to
secretory antigens and a major allergen of the nematode parasite Ascaris. The
Journal of Immunology, 143, 2349-2356.
Uhlar, C.M., Burgess, C.J., Sharp, P.M., Whitehead, A.S., 1994, Evolution of the
Serum Amyloid A (SAA) Protein Superfamily., Genomics 19, 228-235.
Vander, A.J., Sherman, J.H., Luciano, D.S., 1994, From Human Physiology (The
Mechanisms of Body Function) International Edition, 6th Edition, McGraw
Hill Inc.
Vautier, H.E., and Postigo, C.B., 1986, Bovine Mastitis and antiobiotic residues in
milk: risks to public health., World animal review 60, 41-43.
Versalovic, J., K apur, V., Mason, E.O. Jr., Shah, U., Koeuth, T., Lupski, J.R.,
Musser, J.M ., 1993, Penicillin-Resistant Streptococcus pneumonia Strains
Recovered in Houston: Identification and Molecular Characterization of
Multiple Clones., The Journal of Infectious Diseases, 167, 850-856.
Versalovic, J., Koeuth, T., Lupski, J.R., 1991, Distribution of repetitive DNA
sequences in eubacteria and application to fingerprinting of bacterial
genomes. Nucleic Acids Research, 19,24,6823-6831.
Wang, G., W hittam, T.S., Berg, C.M., Berg, D.E., 1993, RAPD (arbitrary primer)
PCR is more sensitive than multilocus enzyme electrophoresis for
148
distinguishing related bacterial strains. Nucleic Acids Research, 21, 5930-
5933.
Watson, D.L., 1976, The effect of cytophilic IgG2 on phagocytosis by ovine

polymorphonuclear leucocytes. Immunology, 31, 159.
Watson, D.L., Lee, C.G., 1978, Immunity to experimental staphylococcal mastitis-
comparison of live and killed vaccines. Australian Veterinary Journal, 54,
374-378.
W ebster, J., 1993, Understanding the dairy cow. Second Edition, Blackwell
Scientific Publications.
Wedmore, C.V., Williams, T.J., 1981, Control of vascular permeability by
polymorphonuclear leucocytes in inflammation. Nature, 289, 646-650.
Whicher, J.T., Dieppe, P.A., 1985, Acute phase proteins. In Clinics in Immunology
and Allergy, 5, 425-46. Laboratory Investigation of Immunological
Disorders, ed. R.A. Thompson.
Whitehead, A.S., De Beer, M.C., Steel, D.M., Rits, M., Lelias, J.M ., Lane, W.S.,
DeBeer, F.G , 1992, Identification of novel members of the serum amyloid A
protein superfamily as constitutive apolipoprteins of high density lipoprotein.
Journal of Biological Chemistry, 267, 3862-3867.
Wilkins, J., Gallimore, J.R., Moore, E.G., Pepys, M.B., 1993, Automated
immunoassay for SAA. Abstract of VII International Symposium on
Amyloidosis, 60.
Williams, R.C., Gibbons, R.J., 1972, Inhibition of bacterial adherance by secretory
immunoglobulin A: A mechanism of antigen disposal. Science, 177, 697-
699.
Williams, M.R., Hill, A.W., 1982, A role for IgM in the in vitro opsonisation of
Staphylococcus aureus and Escherichia coli by bovine polymorphonuclear
leucocytes. Research in Veterinary Science, 33, 47-53.
Williams, J.M ., Mayerhofer, H.J., Brown, R.W., 1966, Clinical evaluation of a
Staphylococcus aureus bacterin (polyvalent somatic antigen). Veterinary
Medical Small Animal Clinics, 61, 789-793.
149
Williams, J.M., Shipley, G.R., Smith, G.L., 1975, A clinical evaluation of
Staphylococcus aureus bacterin in the control of staphylococcal mastitis in
cows. Veterinary Medical Small Animal Clinics, 70, 587-594.
Wilson, C.D., 1981, Mastitis control and herd management. Eds Bramley, A.J.,
Dodd, F.H. and Griffin, T.K., Proc. Nat. Inst. Res. Dairy reading UK
Wolcott, M .J., 1992, Advances in Nucleic Acid-Based Detection Methods. Clinical
Microbiology Reviews, 5, 370-386.
Wood, A.C., Todd, I., Cockayne, A., Arbuthnott, J.P., 1991, Review.
Staphylococcal enterotoxins and the immune system. FEMS Microbiology
Zola, H., 1995, Monoclonal Antibodies, The Second Generation. Bios Scientific
Publishers Limited.
| GLASGOW
UNIVERSITY
LIBRARY
150

Infection and Inflammation in Bovine: Staphylococcus Au Reus Mastitis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Infection and Inflammation in Bovine: Staphylococcus Au Reus Mastitis

Uploaded by

Copyright:

Available Formats

INFECTION AND INFLAMMATION IN BOVINE

STAPHYLOCOCCUS AUREUS MASTITIS

FIONA JANE YOUNG

For the Degree of MASTER OF VETERINARY SCIENCE

Department of Veterinary Clinical Studies

Fiona Jane Young, 1997

All rights reserved

INFORMATION TO ALL USERS

In the unlikely e v e n t that the a u thor did not send a c o m p le te m anuscript

Published by ProQuest LLC(2018). C opyright of the Dissertation is held by the Author.

All rights reserved.

To investigate the acute phase response as an indicator of inflammation in mastitis,

1.1 Common staphylococcal species and their h osts.................... 16

2.9 Staphylococcus aureus strains isolated from each quarter of

2.1 REFP of bacterial isolates from cow 186 on day -4 (pre-challenge)

I am indebted to Dr. Julie Fitzpatrick for her supervision, enthusiastic

I wish to express my gratitude to:

Finally, I would like to acknowledge the University of Glasgow James Houston

Fiona J. Young, May, 1997

1.1.1 The Disease

Bovine mastitis is a serious welfare and economic problem in the United

Mastitis, defined as inflammation of the mammary glands, produces clinically

1.1.3.1 Contagious Pathogens

Primary udder pathogens, including S. aureus and S. agalactiae, are considered to

Streptococcus dysgalactiae is most frequently transmitted during the milking

1.1.3.2 Environmental Pathogens

Streptococcus uberis is found naturally on mucous membranes, skin and in the

Other environmental organisms which can cause mastitis include Pseudomonas,

1.1.3.3 Minor pathogens

Micro-organisms classed as ‘minor pathogens’ are usually considered to be non-

1.1.4 Udder Defences

Non-specific defences include lactoferrin, lysozyme and lactoperoxidase and

The percentage of lymphocytes in cells from mammary secretions are relatively

Cells of the monocyte-macrophage series are the major type in mammary

In spite of the presence of considerable numbers of immune cells in the local

1.1.5 Treatment and Control

1.2 The Staphylococci

Members of the genus Staphylococcus are facultatively anaerobic, non-motile,

Table 1.1: Common staphylococcal species and their natural hosts

1.2.2.1 The Staphylococci

A number of factors have been shown to be important in the pathogenicity of

Several other substances which are involved in resistance to phagocytic ingestion

1.2.2.2 Staphylococcus aureus

Staphylococcus aureus is a particularly well adapted pathogen for the mammary

The pathogenic capacity of a particular strain of S. aureus is due to the combined

The initial line of defence against staphylococcal infection is the physiologically

Abscess formation and fibrosis associated with staphylococcal infection reduces

A major limitation in the effective control of staphylococcal mastitis remains its

Species identification of a microorganism is important for diagnosis, treatment

In bacterial typing, the reliability of a typing system depends on the bacterial

Phenotypic procedures take advantage of biochemical, physiological and

1.2.3.2 The Staphylococci

DNA restriction endonuclease fingerprinting (REF) measures restriction fragment

Mastitis is clinically recognisable by pain and inflammation of the udder (Webster,

Inflammation may be initiated in various ways through different pathways (Evans

1.3.1 Acute Phase Response

The APR is controlled by a complex array of cytokines and hormones which

1.3.2 Acute Phase Proteins

In addition to using APP levels as markers of disease in the live animal,

Bovine haptoglobin is a highly polymerised serum haemoglobin binding protein

Haptoglobin concentration in bovine serum was originally expressed as a

1.3.3.2 Serum Amyloid A

The regulation of SAA and haptoglobin appear to be disease specific pathways.

Day -4 (pre BBBBBBBBB

Day -4 (pre AAAAAAAA AAAAAAAAA AAAAAAAAA AAA*AA

Day 3 (post AAAAAAAAA AAAAAAAAA AAAAAAAAA AAA*AA